TRAINING DEVICES FOR TRAJECTORY-BASED SPORTS
Methods and apparatus related to improving player performance for trajectory-based sports are described. In particular, sporting devices are described that can be utilized to improve player performance in basketball. The sporting devices can include a camera-based system configured to capture and analyze the trajectory of a shot taken by a player. The camera-based system can be configured to provide feedback that allows a player to optimize the trajectory mechanics associated with shooting a basketball. In one embodiment, the camera-based system can be used in conjunction with a training aid that is attached to a basketball rim. The training aid can be configured to improve the trajectory mechanics of individuals utilizing the modified basketball rim to practice their shooting.
Latest Pillar Vision, Inc. Patents:
- Systems and methods for tracking basketball player performance
- SYSTEMS AND METHODS FOR DETERMINING REDUCED PLAYER PERFORMANCE IN SPORTING EVENTS
- Trajectory detection and feedback systems for tennis
- Systems and methods for tracking dribbling and passing performance in sporting environments
- Systems and methods for tracking dribbling in sporting environments
CROSS REFERENCE TO RELATED APPLICATIONS
This application claims priority under 35 U.S.C. §119(e) from co-pending U.S. Provisional Patent Application No. 61/286,474, filed Dec. 15, 2010, titled “TRAINING DEVICES FOR TRAJECTORY-BASED SPORTS,” which is incorporated by reference and for all purpose.
This application is related to U.S. patent application Ser. No. 12/127,744, filed May 27, 2008, by Marty, et al, title, “Stereoscopic Image Capture with Performance Outcome Prediction in Sporting Environments,” which is incorporated in its entirety and for all purposes.
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates generally to devices and systems for sports training and entertainment and more specifically to a trajectory detection and feed back systems and outcome prediction in sporting environments.
2. Description of the Related Art
In sports, player performance is primarily results based. A player is said to be a good player when they produce a consistent result over some range of circumstances. For instance, a professional basketball player might be considered good when on average they produce a certain number of points per game, rebounds, assists, etc., over the course of a season. A player is said to be a great player when they produce a consistent result in more extreme circumstances, such as, in a championship or play-off game as well as providing good performances on average at other times. For example, some basketball players are known for being able to “take over a game” or impose their will on another team in certain situations and are considered to be great for this ability.
The difference between a great and a good player is often described as some intangible quality, such as their will or drive to succeed. Sometimes even when a player produces what appears to be a result consistent with a great player, it is argued that the player is not really great and their performance is a result of circumstances, such as having a really great supporting team. Further, in general, it is often difficult, in a quantifiable manner, to classify and distinguish the performances between players of varying abilities or to distinguish between varying performances by the same player, in regards to answering the questions, such as, why is player 1 good while player 2 is average, why does the performance of a player vary so much, what is a quantifiable different between two performances?
The intangible nature of describing in a quantifiable manner the differences between performances in a sporting environment can be frustrating to players, coaches, broadcasters and spectators alike. Players want to be able identify in a quantifiable manner why their own performances vary from one to another or how their performance varies from a better player so that they can improve their performance. Coaches in team and individual sports want this information so that they can help their players improve. In team sports, coaches may want this information as a way to exploit weaknesses possessed by opposing players. Broadcasters and spectators may want this information because it can add to the entertainment value of watching a sport. Further, spectators are also participants in many of the sports they watch, and thus the spectators may want to be able to quantify and compare their own performances as well as compare their performance to the performances of professional players or other participants of the sport in general.
In view of the preceding paragraphs, methods and apparatus are described in the following paragraphs for determining quantifiable differences between performances in a sporting environment that are not strictly results based. The methods and apparatus may include but are not limited to methods and apparatus related to 1) capturing a performance in a sporting environment, 2) analyzing a performance, 3) comparing performances, 4) presenting results obtained from any analyses or comparisons, 5) archiving captured performances, analyses and comparisons and 6) providing simulations of performances using captured and analyzed performance data.
BRIEF DESCRIPTION OF THE DRAWINGS
The invention will be readily understood by the following detailed description in conjunction with the accompanying drawings, wherein like reference numerals designate like structural elements, and in which:
DETAILED DESCRIPTION OF SELECTED EMBODIMENTS
Reference will now be made in detail to representative embodiments illustrated in the accompanying drawings. It should be understood that the following descriptions are not intended to limit the embodiments to one preferred embodiment. To the contrary, it is intended to cover alternatives, modifications, and equivalents as can be included within the spirit and scope of the described embodiments as defined by the appended claims.
The following figures, 1A, 1B and 1C show the effect of the entry angle of the shot on the size of the make/miss zone. In
At larger entry angles, the hoop appears larger to the basketball. Nevertheless, it has been observed the probability of making a shot decreases when the arc of the shot becomes too large. The probability of making a shot decreases for higher arced shots because changes in entry angle to the hoop at larger entry angles, which are associated with higher arced shots, affects the position of the ball as it enters the hoop more than changes in entry angle at lower entry angles associated with lower arced shots. This effect is illustrated and described in more detail with respect to
In the figures, the initial shot conditions are selected that lead to shots with entry angles from 43-57 degrees. Fifteen shots are shown where the entry angle of the shots increases by one degree increments from 43 to 57 degrees. Trajectories are drawn through the center of the basketball. The trajectories start at the point where the shot is launch and end where the trajectory intersects the plane defined by the basketball rim. As the entry angle of the shots increase the arc of each shot also increases.
A comparison can be made between trajectories with entry angles that differ by one degree. For instance, a comparison can be made between the trajectories with entry angles of 44 and 45 degrees, 45 and 46 degrees, 54 and 55 degrees and 55 and 56 degrees, respectively. It can be seen in the figures that the distance between the locations at which the trajectories intersect the basketball rim is much greater between the 54, 55 and 56 degree shots as compared to the 44, 45 and 46 degree shots. Thus, if a first player tries to shoot shots with an average arc of 55 degrees and a second player tries to shoot shots with an average arc of 45 degrees, an error in the entry angle by the first player has a much greater effect on location at which the trajectory of the shot intersects the plane of the basketball rim than the second player. Entry angle errors for the first player trying to shoot at a 55 degree entry angle is much more likely to lead to a missed shot than entry angle errors for the player trying to shoot an arc of 45 degrees. This effect is further illustrated with respect to
The shape of the make/miss zone can vary depending on the release point of the shot.
The difference between the location of the geometric center of the hoop and geometric center of the make/miss zone suggests that it may be beneficial to shoot the ball deeper in the basket behind the geometric center of the basket. It can be seen in
As an example, at 30 degrees entry angle, the geometric center of the hoop is approximately at the boundary of the make/miss zone. Thus, a shot aimed toward the center of the hoop will miss if a shooting error results in the shot being short at all or slightly off-center. If the shooting error results in the shot being long, then margin of error exists for the shot to still go in. The margin of error is much greater for the long shot than for the short shot. If the shot is aimed behind geometric center of the hoop, such as toward the geometric center of the make/miss zone, the shooting error margins for long and short shots are more equally balanced.
It is often observed that as a player gets tired over the course of a game, their shots are often short. If one player is trained to shoot deeper in the basket behind the geometric center of the hoop and another player is trained to shoot at the geometric center of the hoop, then as both players tire and their shots are more likely to go shorter, the player trained to shoot deeper in the basket may have an advantage over the player trained to shoot at the geometric hoop center. The player that is trained to shoot deeper in the basket may have an advantage over the other player because the player trained to shoot deeper would have a greater margin of error on their short shots than the player trained to shoot at the geometric center of hoop.
The strategy above assumes that for a given player the odds of making a shooting error, such as ‘long or short’ and ‘left or right,’ are equally likely for a given player. With an equal distribution among the probability of making a shooting error, it may be beneficial to balance the error margins for each type of shooting error that can occur. For instance, the player has a similar margin of error for a shooting error that results in the shot being short of its intended target or for a shooting error that results in the shot being past its intended target, i.e., long. If it were demonstrated for a particular player that their shooting errors were biased in a particular manner, such as the player was much more likely overshoot the target they were aiming for rather than undershoot their shot, it might be more advantageous to utilize another shooting strategy for this type of player.
As an example of a shooting strategy for a player with an unequal shooting error distribution, for a player that always tends to shoot their shots long, it may be beneficial to implement a shooting strategy that favors increasing the error margins for long shots as opposed to short shots, such as training to aim their shots more towards the center of the hoop. Conversely, if a player tended to make more shooting errors short, than for this type of player it may be advantageous to devise a shooting strategy where they are trained to aim deeper in the basket as compared to a player equally likely to miss a shot short or a long shot. Similarly, a player that tends to have more shooting errors to the left as opposed to shooting errors to the right might gain an advantage if they trained to aim for a point in the basket that is slightly right of the center of the hoop rather than straight on.
Besides player variations, a likely hood of a shooting error may be affected by shooting physics. As described above, it takes more force to shoot a shot longer than a shot shorter. As the force used to generate a shot increases, velocity errors can become more important. Thus, for a given target location within the hoop, the error distribution for long shots, i.e., past the target location, as a function of the distance from the target location may differ with distance differently than short shots as a function of distance from the target location. For instance, the error rate as a function of distance for long shots may increase faster than the error rate as a function of distance for short shots because of the increasing forces associated with long shots. Based upon the error distribution, a target location can be selected that optimizes their chance of making a shot. As described above, this target location may be different than the geometric center of the hoop and may be closer to the geometric center of the make/miss zone but may not be exactly equal to the geometric center of the make/miss zone.
In general, shooting strategies can be developed for players in general or tailored to an individual player. A general shooting strategy can be developed based on an error analysis of the types of errors players as a group tend to make and then a shooting location within the hoop, which accounts for the make/miss zone, can be tailored that optimizes their chances of making shots based upon an error distribution of shooting errors that the group is likely to make. The optimal target location within the hoop that maximizes their chances of making the shot may be different than the geometric center of the hoop. For instance, the optimal target location may be closer to the geometric center of the make/miss zone, which tends to be deeper in the basket, i.e., behind the geometric center of the hoop. In addition, if because of a player's shooting habits or particular physique, i.e., such as being left handed or right handed, the player's tendency to make shooting errors is biased in a particular direction, then a shooting strategy including a shooting target location can be developed that accounts for the particular shooting errors of the individual. These shooting errors may be determined from analysis of shot data gathered using the measuring devices described with respect
A possible triangular cross section for the training aid is shown in the figure. The height of the triangle can be selected to encourage a player to shoot deeper in the basketball, such as 2-3 inches behind the geometric center of the hoop. Typically, the height of the training aid will be less than 3 inches and possibly 2-2.5 inches above the top height of the rim. The height of the training aid in some embodiments may be lower. For instance, cross-section of the training aid could be more rectangular with the top of the training aid proximate to the top of the rim. The training aid can encourage players to shoot deeper because some shots that would normally be made will be deflected by the training aid.
As described with respect to
When a shot approaches the basketball hoop along a particular trajectory, it can pass over the front of the rim and then interact with the back of the rim. As described above, a training aid can be developed that encourages a player to shoot deeper in the basketball by adding a hoop insert that blocks shots that just pass over the front the rim along its trajectory within some margin to encourage shots deeper in the basket. For instance, an insert can be designed that works for straight-on shots, i.e., a shot vector that is perpendicular to the backboard. This training aid can extend around some portion of the front of the rim. Since the insert only extends around the front portion of the rim, the back of the rim is unaffected and rim effects can be correctly reproduced.
To allow for non-straight on shots, the insert can be extended around perimeter of the rim. However, if the insert is extended too far, the insert can interfere with rim effects related to the back part of the rim associated with the shot vector. For instance, if the an insert extended around the front half of a basketball hoop, then for a shot from one side of the hoop, i.e., with a vector parallel to the back board, the insert can block shots that just cleared the front rim, which is desired, but can also block shots that would hit the back of the rim and go-in, which is not desired. In one embodiment, an insert can be provided that covers about 90-110 degree arc of the rim, i.e, about one quadrant of the basketball hoop. The insert can extend from about or slightly past the front of the rim to one side of the rim to allow for straight on shots as well as for shots from the side of the rim from one side of the basket only.
In another embodiment, the rim insert could be placed on a track, such that its position can be rotated around the rim. If a motor is provided, this rotation can be provided through actuation of the motor. Else, the rotation could be performed manually, such as by using a pole. As an example, an insert that spanned proximately a quarter of the hoop could be rotatable from one side of the hoop to another side of the hoop. To allow for shots from one side of the hoop, the insert would be rotated to a first position. To allow for shots from opposite side of the hoop, the insert can be rotated to the opposite side of the hoop by sliding it around the rim.
In yet another embodiment, if the insert can include parts that are rigid or flexible depending on the direction of the shot. For example, an insert can be configured that spans across the front-half of the rim. For a shot from one side of the basket, the part proximate to the front of the rim can be rigid, while the part proximate to the back of the rim can be flexible so as not to prevent the ball from hitting the back rim and going in. For a shot from the other side of the basket, the part that was previously flexible can be designed to be rigid as it is now the front of the rim for this shot and the part that was previously rigid can be flexible as it now the back of the rim for this shot. In one embodiment, the flexibility and rigidity of the insert can be varied using sensors and mechanical actuators.
The training aid can be tailored to shooting errors associated with a particular player, such that it encourages a shooting target location that is optimized for the shooting errors of an individual. As an example, the training aid may be asymmetric around a line drawn between a shot location and the center of the hoop to encourage a shot target location that is left or right of this line for a straight on shot. As described above, this may be advantageous if the player shooting errors are biased to the left or right of this line in some manner. For instance, for a player that tends to make more shooting errors to the left as opposed to the right, the training aid may extend deeper into the rim on left side of the hoop as opposed to the right to encourage the player to target a shot location that is shifted to right of center for a straight on shot. Shooting slightly to the right of the center of the hoop may better accommodate the player's tendency to the make more shooting errors to the left.
A trajectory capture and feedback system, as is described in the following figures, can also be used to help a player train to shoot to a target location in the hoop that differs from the geometric center of the hoop. This system could be used a training device as described with respect to
The basketball hoop 103 may be mounted to a backboard 151 with a support system to hold it up, such as a pole anchored into the ground, a support anchored into a wall or supports suspended from a ceiling. The basketball hoop 103 may be of a standard height and the basketball may be a standard men's size basketball. However, trajectories for a basketball of a different size, such as a women's ball, shot at basketball hoop of varying heights may also be detected and analyzed with the present invention.
The camera 118 in the machine vision system records physical information within a detection volume 110. The physical information that is recorded is images of objects at a particular time in the detection volume 110. The images recorded at a particular time may be stored as a video frame 106. The camera 118 may capture images of the basketball 109 as it moves in trajectory plane 104 as well as images of other secondary objects. The secondary objects may be closer to the camera than the basketball 109 (i.e., between the camera 118 and the trajectory plane 104) or the secondary objects may be farther away from the camera than the basketball 109 (i.e., beyond the trajectory plane 104). The machine vision system may utilize software to distinguish between the movement of secondary objects that may be detected and the movement of the basketball 109.
The trajectory detection system 100 may be set-up in a playing area where basketball is normally played, such as a basketball court with playing surface 119 located in gymnasium or arena. The system 100 may be positioned on the side of court and remotely detect the trajectories of the shots by shooter 112 using the machine vision system. Thus, the shooter 112 and defender 114 may engage in any of their normal activities on the playing surface 119 without any interference from the detection system 100. In the figure, the shooter 112 is guarded by a defender 114. However, the system 100 may also be used when the shooter 112 is unguarded.
With a machine vision system that uses a single camera 118, the locations where the trajectory 102 may be accurately analyzed may be limited. In one embodiment, with the set-up of the trajectory detection system 100 on the side of playing surface 119, accurate analysis may require that the shooter 112 shoot from within the active area 108. In this alignment, the trajectory plane 104 may be nearly normal to the basketball backboard 151. Although, the system 100 may accurately detect and analyze trajectories where the angle between the trajectory plane 104 and the normal to the backboard 151 is within a few degrees. The active area 108 may be different for different systems 100. Further, the present invention is not limited to machine vision systems for detecting the trajectory of the basketball and other sensor systems may allow for different active areas.
The trajectory system 100 may be set-up in different locations around the playing surface 119. By moving the system 100, the active area 108 may be changed. For instance, the trajectory detection may be positioned behind the backboard 151. For this set-up, the active area 108 may be a rectangular area on the playing surface 119 that is parallel to the backboard 151.
Although the active area 108 may be limited with a single camera 118 in a machine vision system, an advantage of the system is it simple to set-up and to operate. With some multiple camera machine vision systems, the active area may be larger than with a single camera system. However, the set-up and calibration of a multi-camera system may be more time consuming as compared to a single camera system because a known alignment of the cameras relatively to one another and relative to the tracked object is needed to process the data.
The single camera system 100 is simple enough to be capable of autonomous set-up and operation with minimal user input. The system may autonomously calibrate itself using known distance markers, such as the height of the basketball hoop or a distance to a free throw line or 3-point arc, which may be captured in video frame data. In another embodiment, a user may be required to stand within the detection zone of the system, holding a basketball or other object, at a fixed distance from the camera and at a fixed height. After the system is calibrated, a user may use the system 100 to practice without the help of an additional operator to run to the system 100. The system 100 may accept voice commands allowing the user to adjust the operation of the system from a distance.
To analyze a trajectory 102 of the basketball 109, the camera 118 may record a sequence of video frames in the detection volume 110 at different times. The number of frames recorded by the camera over a give time period, such as the duration of the ball's trajectory 102, may vary according to the refresh rate of camera 118. The captured video frames may show a sequence of states of the basketball 109 at different times along its trajectory 102. For instance, the camera 118 may capture 1) an initial state 105 of the trajectory shortly after the ball leaves the shooter's hand, 2) a number of states along the trajectory 102, such as 120, 121, 122 and 123 at times t1, t2, t3 and t4 and 3) a termination point 107 in the basketball hoop 103. Although not shown, the system may also be used to generate parameters for characterizing the trajectory of missed shots relating to the rebound flight path, such as but not limited to a rebound height, rebound angle, rebound velocity.
The sequence of captured video frames may be converted to digital data by a video capture card for analysis by the CPU 116. The analysis of video frame data may require the detection volume 110 to remain constant during the trajectory 102. However, the detection volume 110 may be adjusted to account for different set-up conditions of a playing area where the system 100 is employed. For instance, the camera 118 may be capable of zooming in or out of a particular area and changing its focus.
The series of frames used to capture the trajectory may also capture the shooter 112 shooting the basketball 109 including all or a portion of the shooter's 112 body as well as the defender's body 114 during the shot. The physical information captured by the camera 118 regarding the shooter 112 and the defender 114 may also be analyzed by the system 100. For example, different motions of the shooter 112 may be analyzed by the system 100 determine if the shooter is using proper shooting mechanics. As another example, data, such as, a jump height, hang-time, a release point floor position on the playing surface 109, a landing position on the playing surface 109 may be determined using the video frame data captured by the camera 118 in the machine vision system.
After detecting and analyzing the trajectory 102, the system 100 may generate one or more trajectory parameters. The one or more trajectory parameters may be output as feedback information to the shooter 112 and the defender 114. Typically, the system 100 may provide the feedback information while the shot is in the air or shortly after the shot has reached the hoop 103. The feedback information may be provided within less than a second or less than 10 seconds of the initiation of the shot depending on the type of feedback information that is generated. The immediate feedback may increase the training benefits of using the system. The shooter 112 may use the feedback information to improve their skill at making shots. The defender 114 may use the feedback information to improve their defense in preventing the shooter from making their shots. A brief description of the methods used to develop the feedback information is described as follows.
The shooter 112 may also use the feedback information for rehabilitative purposes. For instance, after an injury and/or for psychologically reasons, a player's skill at shooting may decline from a previously obtained skill level. In rehabilitative setting, the present invention may be used by the player to regain their previous skill level and even improve upon their previous skill level. For instance, the feedback information provided by the present invention may increase a shooter's confidence which may provide psychological benefits that lead to an improvement in performance.
To develop basketball feedback information, the basic nature of a basketball shot is considered with the objectives of 1) informing the player in regards to what are a set of optimal trajectory parameters that they can adjust to increase their probability of making a shot and 2) informing their player about how their shots compare to the optimal. This information is output to the player as feedback information. As an example of this process, the basketball shot by the shooter 112 is described. However, the system 100 may be applied to the trajectories of other objects in different sports where optimal trajectory parameters may be different than basketball. Thus, the description is presented for illustrated purposes only.
The basketball shot by the shooter 112 travels in an essentially parabolic arc in the trajectory plane 104. The arc is essentially parabolic and the ball 109 travels in-plane because after the ball is released the dominant force acting on the ball is gravity 109. Other forces, such as ball spin, or if the ball is shot outside, wind, may cause the trajectory to deviate from a parabolic arc. But, when the ball is shot inside, these forces cause little deviation from the parabolic trajectory and a parabolic arc is a good approximation of the trajectory 102.
For each shot by the shooter with an initial release height, there are many different combinations of release velocity and release angles at the initial state 105 that allow the player to make the shot, i.e., the ball travels through the basket 103 and then many combinations of release velocity and release angles where the player does not make the shot. When a player shoots the basketball 109, the player selects a combination of release velocity and release angle. Typically, the selection of the shot parameters is performed intuitively and the player doesn't consciously think of what release velocity and release angle they are selecting. However, through training, the player may be able to improve their intuitive shot selection.
Within the group of the different combinations of release velocity and release angle that may be selected by the shooter, there are combinations of release velocity and release angle that provide the shooter with a greater or lesser margin of error for making the shot. For instance, for a basketball shot in the basket 103, an optimal entry angle into the hoop that provides the greatest margin of error is about 43-45 degrees measured from a plane including the basketball hoop 103. These optimal trajectories are close to trajectories that allow for the ball to reach to the basket 103 with a minimal amount of energy applied by the shooter. For perturbations around this optimal entry angle, such as when the defender 114 causes the shooter 112 to alter their shot, there are more combinations of release velocity and release angle that allow the shot to be made as compared to other combinations of release velocity and release angle away from the optimal.
With the general understanding of basketball trajectories provided above, methods may be developed for providing feedback information that allows for the shooter 112 to train for an initial state 105 that provides the greatest margin of energy i.e., a near minimum energy trajectory. In one embodiment of the present invention, an entry angle and an entry velocity of the basketball 109 near the termination point 107 are two trajectory parameters that may generated from the physical information recorded by the machine vision system in system 100. The entry angle and entry velocity are correlated to the release angle and the release velocity of the shot 102. Thus, after the shooter 112, releases the shot, the camera 118 may record a series of video frames with images of the ball 109 as it approaches the basket 103. With this information, the entry angle and the entry velocity of the shot may be generated. One or both of these trajectory parameters may be provided to the player as feedback information.
The feedback information may be provided to the shooter 112 and the defender 114 in one of a visual format, an audio format and a kinetic format. For instance, in one embodiment, on a visual display, the entry angle and/or entry velocity may be viewed in a numeric format by the players, 112 and 114. In another embodiment, when projected through an audio device, numeric values for these parameters may be heard by the players, 112 and 114. The audio feedback device may be a speaker built into the system 100, a speaker connected to the system 100 or audio devices worn by the players, 112 and 114 that receive information from the system 100. In yet another embodiment, a kinetic device, such as a bracelet or headband worn by the players may be used to transmit the feedback information in a kinetic format. For instance, the bracelet may vibrate more or less depending on how close the shot is to the optimum or may get hotter or colder depending on how close the shot is the optimum. Multiple feedback output mechanisms may also be employed. For instance, the feedback information may be viewed in a visual format by coaches or other spectators on a display while a sound projection device may be used to transmit the feedback information in an audio format to the players.
In general, the parameters may be presented qualitatively or quantitatively. An example of qualitative feedback may be a message such as “too high” or “too low” in reference to the entry angle of a shot by the player or “too fast” or “too slow” in reference to the entry velocity. An example of qualitative feedback may be the actual entry angle or entry velocity of the shot in an appropriate unit of measurement, such as a message of “45 degrees” for the entry angle. Again, the qualitative and/or quantitative information may be presented in different formats, such as a visual format, an auditory format, a kinetic format and combinations thereof.
With knowledge of what are optimal values of the trajectory parameters transmitted in the feedback information, the shooter 112 may adjust their next shot to generate a more optimal trajectory. For instance, if the feedback information is an entry angle and their shot is too flat, then the shooter 112 may adjust their next shot to increase their entry angle. Conversely, with their knowledge of what are the optimal values of the trajectory parameters, the defender 114 may adjust their defensive techniques to force the shooter 112 to launch a shot along a less than optimal trajectory 102. Thus, the defender 114 can experiment with different techniques to see which are most effective. In different training methods, the system 100 may be used to measure a trajectories for a shooter 112 training without a defender 114 or as is shown in the figure training with the presence of a defender 114.
The feedback information may be provided to the player before prior to the ball 109 reaching the basket or shortly after the ball reaches the basket 103. The system 100 is designed to minimize any waiting time between shots. For each shooter and for different training exercises, there may be an optimal time between when the shooter shoots the ball 109 and when the shooter 112 receives the feedback information. The system 100 may be designed to allow a variable delay time between the shot and the feedback information to suit the preferences of each shooter that uses the system 100 or to account for different training exercises that may be performed with the system. For instance, a rapid shooting drill may require a faster feedback time than a more relaxed drill, such as a player shooting free throws.
The present invention is not limited to providing feedback information for near minimum energy basketball trajectories. For instance, under some conditions, such as when a smaller player shoots over a larger player, it may be desirable for the shooter to shoot with a greater than optimal arc to prevent the larger player from blocking the shot. Thus, the shooter may use the feedback information provided by the system 100 to train for different conditions that may call for different types of shots, such as shooting over a larger player as compared to a wide-open shot. Further, the trajectory analysis systems of the present invention may be used to train in different types of basketball shots, such as bank shots, hook shots, lay-ups, jump shots, set-shots, free throws and running shots, that may requiring the mastery of different shooting skills and may have different optimal trajectory parameters. Thus, the detection system 100 may be adjustable to allow for training in different types of shots. Further, for different sports, different trajectory skills may be optimal for improving performance, which may be different than basketball. The different trajectory skills that may be required for different sports may be accounted for in the present invention.
A measure of how good a player's shooting skills may be a consistency of their trajectory parameters averaged in some manner over many shots. Typically, it has been determined empirically that better shooters have a lower variability in their trajectory parameters for a given shot, such as a free throw. Thus, to rate a shooter's performance, it may be desirable to generate trajectory parameters for a plurality of trajectories shot by a player in a trajectory session and then calculate a standard deviation for each of the trajectory parameters.
The standard deviation (SD) is a measure of the scatter of a particular set of data. It is calculated as,
where ymean is an average value of trajectory parameter, N is the number of trajectories and yi is a value of the trajectory parameter for a particular trajectory. There are other types of statistical parameters that may be used to characterize data variability and the present invention is not limited to the standard deviation formula described above.
During a trajectory session where a plurality of trajectories are analyzed by the system 100, the trajectory parameters generated for the plurality of trajectories may be stored to a mass storage device contained in the system 100 or in communication with the system 100. After the session, the standard deviation for all the trajectories in the session may be generated. In other embodiments, to provide measures of variability of different data sets representing different playing conditions, the system 100 may divide the trajectory data into different subsets, such as grouping according to types of shots, locations of shots, shots where the shooter is guarded, shots where the shooter is unguarded, made shots, swished shots, missed shots, shots made earlier in the session versus shots made later in the session, and combinations of these groupings.
The statistical variability calculated from the different data sets may be used as a guide by the system for suggesting methods that will improve the player's shooting skills. The system 100 may include software for suggesting methods based upon the statistical analysis. For instance, the system 100 may determine that a player's shot variability is greater when they are guarded as opposed to unguarded, thus, exercises may be prescribed to the player that focus and shooting while guarded. As another example, the player's shot variability may be greater later in a session as opposed to earlier in a session or greater in a training session before practice as opposed to after practice, thus, the system may suggest the player work on their aerobic conditioning. In yet another example, the player's shot variability may vary as a function of a distance from the basket and the system may suggest the player concentrate on shots at the distances where the variability is greatest.
In some embodiments, the trajectory session data and other information generated by the system 100 may be viewed via a number of different output mechanisms, such as a hard copy from a printer or a display. For example, a printer connected to the system 100 may be used to generate print-outs of trajectory session data in different formats. As another example, a display interface in communication with the system 100 may be used to view trajectory session data in different formats. In particular, the system 100 may include a touch screen interface for viewing trajectory session data and providing input parameters into the system. As another example, the system 100 may communicate with a portable viewing device capable of interfacing with the system 100.
Information generated with system 100, such as trajectory data from a plurality of trajectories in a trajectory sessions, may be archived. The archival storage system may be a remote storage device in communication with the system 100 or may be a mass storage device provided with the system 100. The archival storage system may include raw data of physical information recorded by the camera 118, such as video frame data, as well as, trajectory parameters and other information generated from analysis of the raw data. The archival data may store trajectory session data for a plurality of different trajectory sessions by one or more different players.
By accessing the archival data, an improvement over time for a particular parameter generated by the system 100, such as a shot variability parameter, may be assessed. Further, the archival data may be used for data mining and video editing purposes. For instance, in a video editing application, the graphic of the player's average trajectory may be integrated with video data of the player shooting. In another example, video clips of two or more different players shooting may be compared or video clips of a single player shooting during different trajectory sessions may be compared to show the player's improvement. In data mining applications, the video data may be further analyzed to characterize a player's shot mechanics. In another application, simulations may be generated to predict gains in team performance based-upon improvements in individual performance on the team. This type of simulation may require archival trajectory session data to be analyzed for a plurality of different players.
In some embodiments, the archival data may be accessible via a remote connection. For instance, a password-protected web-site may be used as a portal for accessing archival data generated from system 100. The web-site may allow clients, such as players, coaches, or scouts to gain access to the web-site from remote sites, such as home computer connected to the Internet or a portable computer connected to the Internet. The web-site may include a plurality of analysis tools and a graphical interface for viewing graphical data from the applications in different formats. In another embodiment, the archival data may be downloaded to a CD, DVD or other portable storage medium that the player can take with them. Analysis software may also be downloaded with the archival data so that the player can analyze the data on another computer.
Information generated during a trajectory session may be stored in a database. The database may relate player identification information, such as a name, an address, a team, a session time, a session location, a session data to raw data recorded during the trajectory session and information generated during the trajectory session. The database may be used for player tracking purposes and targeting services to players that have used the trajectory system.
When the device is turned on, it may be configured to begin providing feedback information until it is turned off. The device may include a wireless or wired remote to provide options, such as but not limited to changing the feedback format options, recalibrating or resetting the device, changing the volume of a speaker or turning the device on or off. In one embodiment, the device may include a limited memory that can be overwritten over time. The memory may include captured images and/or associated trajectory analysis. For instance, the memory can be sized to store 24 hours of video that includes a time stamp. After the 24 hours of video are filled, the device can begin to rewrite over the oldest stored video with new captured video data.
The device 100a may include a wired and/or wireless interface that allows stored data including image data to be downloaded to a separate device, such as but not limited laptop computer or a cell phone. For instance, image data could be downloaded over a particular time period as selected by a user and downloaded to their device. The device 100a may provide video indexing and skipping features that allow a user to locate video data stored on the device and review it on their device. For instance, a preview capability could be provided that is compatible with a user's cell phone, such that the video stored on the device 100a could be viewed on a screen associated with the user's cell phone. Software downloads could be provided from the device 100a or from another remote location, such as server associated with a website, that allows captured image data downloaded from the device 100a to be further manipulated and stored on the user's device, such as their cellphone or laptop. In another embodiment, a web-site may be set up that allows data stored on their personal device, such as a cell phone or laptop, to be uploaded an further processed using software provided in association with the web-site
In other embodiments, a server associated with a web-site, may be configured to download software to a user's device with a camera that allows it to behave like the machine vision system described with respect to
Thus, when one of these devices where placed near a sports venue, such as indoor or outdoor basket ball hoop structure, the device could be used to provide image capture and trajectory analysis capabilities. If the device included audio capabilities, it could also be used to provide feedback information. For instance, a cell phone could be configured to communicate with a Blue-Tooth™ capable head set worn by a user and coupled to the cell phone to provide feedback information. In another embodiment, a portable speaker system could be coupled to the cell phone to provide audio capabilities.
In one embodiment, a docking station or stand, such as a tripod, could be provided for use with a user's device, such as a cell phone, camera or camcorder. The docking station could be configured to allow a user's device to be fixed in a particular orientation. For instance, the docking station or stand could include a flexible device that attaches to the cellphone or camera that allows it to be mounted in various orientations, such as a multi-armed Gorrilapod.™ The docking station may also be configured to allow augment features of the user's device. For instance, the docking station could include a power source, a power/communication interface, such as a USB compatible interface, additional electronic storage and additional processing.
The docking station could also be designed with an enclosure that provides some protection to the device. For instance, the enclosure could cover the device to provide some protection from rain. Further, the enclosure could partially enclose and secure the user's device to protect it from damage if it were impacted in some way, such as hit by an errant basketball or other object for which the device is being used to capture and analyze trajectory data.
Calculating Basketball Trajectory Dynamics: Basketball Swish/Make Analysis
When a basketball shot goes through the hoop without contacting any part of the rim, we call it a “swish.” Since the hoop is larger than the ball, there can be some variation in the shot and still get a swish. This analysis first determines the range of straight-on shots that will swish. The shot possibilities may be described by locating the center of the ball at the instant it passes through the hoop or, more technically, when the center of the ball lies in the plane of the hoop. The locus of points that describe shots that swish may be referred to as the swish zone, and can be illustrated by drawing a top view of the hoop with the swish zone defined within that circular region. The reference point (coordinate system origin) is the center of the hoop, with the positive x-axis projecting to the right and the positive y-axis projecting toward the back of the hoop, as shown in
For the purposes of illustration, it may be assumed that the hoop is 9″ in radius and the ball is a men's pro ball, approximately 4.77″ in radius. These assumptions may be varied to address different hoop and ball geometries as desired. For this example, it may be assumed that the ball's trajectory is a straight line in the immediate vicinity of the hoop, described by the hoop entry angle and measured from the horizontal. In this frame of reference, a perfectly flat shot would be 0 degrees and a ball dropped from directly above the hoop would be 90 degrees. This analysis may be generalized to include a more representative parabolic trajectory, or even a trajectory corrected for aerodynamic and buoyant forces, if necessary. For this example, it may be assumed the curvature of the trajectory in the immediate vicinity of the hoop is insignificant.
For the ball to swish, it must pass over the front of the rim and under the back of the rim. These constraints limit the range of possible shots that may swish for a given entry angle, and the widest part of the ball (its diameter) determines the extent of that range. The geometry may be described for the longest possible swish at a particular entry angle and shortest possible swish at a particular entry angle.
As can be seen in
y=location of center of ball in plane of hoop
Rhoop=radius of basketball hoop (set at 9″)
Rball=radius of basketball (set at 4.77″)
θ=shot entry angle measured from horizontal
This formula results in the following values for the longest and shortest possible swish shots as a function of shot entry angle:
These results apply to a straight-on shot. In order to calculate the swish zone for off-center shots, compound angles must be computed that are difficult to illustrate geometrically. The result is more easily depicted graphically by showing the swish zone within the hoop. For example, the swish zone for a 45-degree shot is shown in
It is still possible for the ball to go through the hoop if it does not pass through the swish zone. Shots near the swish zone will hit the rim. Depending on the shot and hoop geometry, it is possible for the ball to bounce off the rim and pass downward through the hoop. The locus of points for which this occurs may be called the make zone. The boundary of the make zone may be defined by the longest shot that can hit the back rim and rebound downward to the plane of the hoop before contacting the front rim.
The location of the boundary of the make zone for a straight-on shot may be determined from the following system of equations, which are shown in
h=height of center of ball above plane of hoop when ball strikes back rim
rb=radius of ball
rh=radius of hoop
y1=y-location of center of ball in plane of hoop when ball strikes back rim
β=angle between rim-normal and plane of hoop
γ=angle between plane of hoop and resultant ball direction after rebounding off back rim
These equations do not have a closed-form solution. Nevertheless, the equations may be solved iteratively using a variety of progressive solution techniques, such as the Newtonian method. Using this method, the location of the make zone boundary may be determined as a function of entry angle. The results of a sample calculation are given in the table below.
The table indicates that below 25 degrees entry angle, there is no make zone. The geometry for off-center rim shots is more complicated, but the make zones may be depicted graphically. There are some idealizations used in generating these plots that may be relaxed if necessary to produce a more accurate depiction of the make zone.
In performing this analysis, a conservative assumption has been made that only rim bounces that cause the ball to continue downward have the potential to score. In practice, many shots are observed to bounce up off the rim, then either drop through the hoop or make subsequent contact with the rim and/or backboard ultimately resulting in a made shot. As a result, the make percentage calculated by the present method may be lower than what would be expected in practice. This observation has been confirmed with empirical data as is described with respect to
One means of addressing this under-prediction of made shots may be to develop a calibration factor based on empirical evidence. The make zone may be enlarged by a multiplicative factor such that the calculated made shots more nearly replicate observations. This factor, which may be dependent on a number of variables such as entry angle and speed, may be deduced from an experimental dataset, then applied to the algorithm for use in subsequent predictions.
Another approach to developing a more accurate make zone may be to incorporate a more complete kinetic model of the ball and its interactions with the hoop, bracket, and backboard. Effects to be included may include the angle of rebound when the ball contacts a solid surface, the effects of spin and ball friction against a solid surface, energy loss of the ball as a result of its impact, and the flexibility and energy absorption of the hoop and associated hardware. This would allow the calculation to model multiple ball/rim interactions and thereby determine the make zone even for very complex dynamics. The bounce dynamics may also be verified experimentally.
A rear view of an experimental set-up for generating basketball trajectories is shown with respect to
The machine 1302 may be programmed to execute a large number of shots and the outcome for each shot may be recorded. The release height of the device may be set to a particular value. From shot to shot, a left-right orientation of the shot relative to the center of the basketball hoop, a release velocity and a release angle may be automatically adjusted. Each of the values, left-right orientation, release velocity and release angle may be varied to generate a number of shots, such as but not limited to 10,000 shots. Different ball sizes may be employed with the machine 1302, such as a men's ball or a women's ball.
The shooting machine may be positioned at a location on the court 1314 that is a fixed distance from the basketball hoop. In this example, the shooting machine 1302 is positioned to generate a shot near the free throw line. The shooting machine is mobile and may be moved to different locations to generate a set of shots at each location.
From an entry angle between 40-65%, the theoretical make/miss percentage is 2-3 percentage points less than the experimentally predicted values. Below 40%, the error between the calculations is greater. As previously, mentioned some rebound dynamics were not considered in the theoretical calculations for the embodiments described herein and hence some shots that may be made are not captured by the methodology, such as hitting the back rim, then the front rim and then going in. At lower entry angles, shots such as these are more prevalent as compared to the higher entry angles and hence there is a larger error in the calculations at lower entry angles.
As previously noted, it may be possible to better model the rim effects and hence produce a more accurate simulation. In another embodiment, it may be possible to correct the theoretical simulation using a correction factor derived from the shooting machine data. In yet another embodiment, it may be possible to generate a series of curves like the one shown in
Of note in
Physical information 216 is input into the system 100 via sensors 212. In one embodiment, a machine vision system may be used where the machine vision system comprises one or more cameras 201 (e.g., a CCD camera) and a video capture card 203 for digitizing captured frame data. The video capture card 203 may capture color pixel data. The camera 201 may employ a 3.5-8 mm zoom lens and may allow for different lens attachments. In another embodiment, the system may employ a plurality of cameras arranged on a mechanism that allows different type cameras to be rotated or moved into place where only one camera is used at a time to record frame data. The different cameras may allow the detection volume of the system to be adjusted.
The digitized frame data from a machine vision system and other sensor data may be processed by a computer 202. The computer 202 may be a modified PC using a 1.6 GHz processor 204 w/RAM and a CD-RW drive 205 for inputting and outputting data and software. The computer 202 may also include a mass storage device, such as hard drive 207 and various network/device communication interfaces, such as wireless and wired network interfaces, for connecting to a local area network (LAN), wide-area network (WAN) or the Internet. The device communication interfaces may allow the computer to communicate with a plurality of peripheral devices and other remote system components.
The computer 202 may include operating system software 206 for controlling system resources, such as feedback interfaces 213 and the system input/output mechanisms 215. The computer 202 may be used to execute analysis software 208 for analyzing trajectories using the sensor data from sensors 212 and for generating feedback information 217. The analysis software 208 may include software for providing various services, such as 1) providing a list or a plot of trajectory session information comprising one or more of physical information, trajectory parameters and feedback information for the plurality of trajectories, 2) comparing the trajectory session information from the trajectory session with trajectory session information from one or more different trajectory sessions, 3) generating trajectory session parameters used to characterize a human's performance in the trajectory session, 4) predicting performance improvement as a function of the trajectory session parameters, 5) prescribing actions for improving performance and 6) performing video editing tasks. The computer 202 may also be used to execute database software for relating physical information 216 and other information generated by the computer 202 to player identification information (e.g., name, age, address, team, school, etc.) and session identification information (e.g., time, data, location, number of trajectories analyzed, types of shots, etc.).
Power to the computer 202 and other devices may be provided from the power supply 209. In one embodiment, the power supply 209 may be a re-chargeable battery or a fuel cell. The power supply 209 may include one or more power interfaces for receiving power from an external source, such as an AC outlet, conditioning the power for use by the various system components. In one embodiment, for in-door/outdoor models, the system 100 may include photocells that are used to provide direct power and charge an internal battery.
Feedback information 217, used by clients of the system 100 to improve their trajectory skills, may be output through one or more feedback interface devices 213, such as a sound projection device 211. In general, the system may be capable of outputting feedback information 217 to a plurality of different devices simultaneously in a plurality of different formats, such as visual formats, auditory formats and kinetic formats.
The system 100 may support a plurality of different input/output mechanisms 215 that are used to input/display operational information 218 for the system 100. The operational information 218 may include calibration and configuration setting inputs for the system and system components. In one embodiment, a touch screen display 210 may be used to input and display operational information 218 using a plurality of menus. Menus may be available for configuring and setting up the system 100, for allowing a player to sign into the system and to select preferred setting for the system 100 and for viewing session information 219 in various formats that have been generated by the system. The printer 214 may be used to output hard copies of the session information 219 for a player or other client of the system 100. The present invention is not limited to a touch screen display as an interface for operational information. Other input mechanisms, such as but not limited, a key board, a mouse, a touch pad, a joystick and a microphone with voice recognition software may be used to input operation information 218 into the system.
Wheels 304 are attached to the chassis 301 to allow the system 100 to be easily moved and positioned for use. In general, the chassis of devices of the present invention may be designed with a weight and a form factor, which may facilitate transport, storage and unobtrusive set-up, calibration and operation of the device. For instance, the device includes a handle 303 attached to panels 300 comprising the housing that may be used to move the device and which may aid in set-up and storage of the device.
The speaker 211 takes up a large portion of the internal volume of the system. In one embodiment, a travel system may be used that incorporates a portable computer system such as laptop that is connected to a machine vision system with the camera 201. To use the travel system, it may be placed on top of a support platform, such as a tripod, a table, a chair or even coupled to a golf bag or golf cart. The travel system may provide feedback information via a wireless communication interface to audio device, such as an “earbud,” worn by the player or wearable feed back device. In another embodiment, the travel system may generate output signals that may be routed through a portable audio system (e.g., a boom box) for amplification via speakers on the audio system to provide feedback information.
The sensor system 502 may convert the physical information 507 to sensor data signals 509. For instance, a charge-coupling device generates electronic signals in response to photons striking a sensor array. The sensor data signals 509 may be sent through a wired or wireless connection to a sensor interface 510, which provides signal conditioning. The signal conditioning may be needed to allow the sensor data 509 to be processed. For instance, prior to analysis, video frame data may be digitized by a video capture card.
In 513, the conditioned signals 511 may be processed according to system control software and according to trajectory analysis software 513 using set-up and control inputs 512 that have been input into the system. The system control software 513 may analyze portions of the data 511 to determine whether the sensor system 502 is operating properly. Based-upon the analysis of the data 511, the system control software may provide calibration instructions and other operational instructions to the sensor system which may be transmitted to the sensors via the sensor interface 510.
The trajectory analysis software 513 may be used to process the conditioned signals 511 and generate trajectory parameters. The trajectory parameters may be used to generate feedback information. The feedback information may be one or more trajectory parameters or a combination of trajectory parameters, such as a ratio of trajectory parameters or a product of trajectory parameters that may be useful to a system client in improving their trajectory skills.
Depending such factors as the application (trajectory of a specific type of object), the set-up and components of the system, the environment in which the system is used and what portion of the trajectory of an object the device is used to measure, the present invention may provide feedback to the player nearly immediately, within a second or within 10 seconds as measured from some time state along the trajectory that has been analyzed by the system. For instance, when information on the beginning of the trajectory is directly generated by the system, then the time to provide feedback may be measured from the time when the trajectory is initiated and then first detected by the system. When information on the end of the trajectory is directly measured, then the time to provide feedback may be measured from the time to when the trajectory has neared completion and has been detected by the system.
The feedback information may be sent as feedback information parameters 516 to one or more device interfaces 517. The device interfaces 517 may communicate with a plurality of feedback devices. The device interfaces 517, which may include device drivers, may transmit device data/commands 518 to a feedback device interface 519 located on each feedback device. The device data/commands 518 may be used to control the operation of the feedback devices. The output from the feedback device may also be modified using set-up/control inputs 520 that may vary for each device.
The feedback devices may output the feedback information parameters 516 received as device data 518 in one of an audio, visual or kinetic format 521 depending on the capabilities of the feedback device. For example, the device interface 517 may send device data/commands 518 to a display that allows a numeric value of a feedback information parameter 516 to be viewed on the display by one of the system clients 522, such as players, coaches and spectators. As another example, a device interface 517 may send device data/commands 518 to an audio output device that allows feedback information parameters 516 to be output in an audio format to one or more of the system clients 522.
The feedback parameters 516 generated from the trajectory analysis software 513 and other raw data generated from the sensor system 502 may be sent to session storage 515. The session storage 515 may accumulate trajectory data from a plurality of trajectories generated during a trajectory session for one or more players. All of a portion of the trajectory data 514 may be sent to archival storage 525 when the session has been completed. For example, only a portion of the raw data, such as video frame data, may be sent to archival storage. Further, the data may be filtered for bad data prior to being sent to archival storage 525. The archival storage 525 may include a database used to relate trajectory data from one or more trajectory sessions to the conditions of the trajectory session, such as time place and location, and player identification information.
The archival data 524 and session data 514 may be used to provide one or more services 523 including but not limited to 1) a session record of trajectory parameters, 2) session diagnostics, 3) prescription for improvement, 4) a history comparison of trajectory data from different sessions, 5) individual/group comparisons of trajectory session data, 6) video analysis and editing tools, 7) simulations (e.g., predicting a player's driving distance improvement based upon changing one or more of their swing parameters and 8) entertainment. As an example of entertainment, a player's trajectory average trajectory parameters and variability may be used in trajectory simulations for a video golf game or another game where the parameters have been measured. Two players that have used the system 100 may both enter their parameters and compete against one another in the video game. The player may also use the game to see how they match up against professional or other athletes who have had their trajectory parameters defined.
Output from the data services 523 may be converted to a portable record 527, such as print-out from a printer, or may be formatted for viewing on a graphical interface 528. The graphical interface may also include a storage capacity allowing data to be viewed at a later time. The output from the data services 523, such as a portable record 527 or information viewed on the graphical interface 528, may be used by the system clients 522. The data services 523 may also be provided via a data mining interface 526. The data mining interface 526 may include analysis tools and a graphical interface. When the archival storage is remotely accessible, it may be used to access archived data 524 via a remote connection, such as from the Internet.
Information passed between the different components in the system may be transmitted using a number of different wired and wireless communication protocols. For instance, for wire communication, USB compatible, Firewire compatible and IEEE 1394 compatible hardware communication interfaces and communication protocols may be used. For wireless communication, hardware and software compatible with standards such as Bluetooth, IEEE 802.11a, IEEE 802.11b, IEEE 802.11x (e.g. other IEEE 802.11 standards such as IEEE 802.11c, IEEE 802.11d, IEEE 802.11e, etc.), IrDA, WiFi and HomeRF.
Although the foregoing invention has been described in detail by way of illustration and example for purposes of clarity and understanding, it will be recognized that the above described invention may be embodied in numerous other specific variations and embodiments without departing from the spirit or essential characteristics of the invention. Certain changes and modifications may be practiced, and it is understood that the invention is not to be limited by the foregoing details, but rather is to be defined by the scope of the appended claims.
1. A system for analyzing a trajectory of a basketball, the device comprising:
- one or more cameras for recording video frame data used to characterize a trajectory of a basketball shot by a human;
- a logic device designed or configured to i) receive the video frame data, ii) generate trajectory parameters that characterize one or more states of the basketball along its trajectory and iii) generate feedback information using the trajectory parameters; and
- one or more feedback output mechanisms for providing the feedback information to the human wherein the feedback information is related to one or more of the following trajectory parameters: 1) an angle of the trajectory relative to a plane of a basketball hoop when the basketball is proximate to the basketball hoop, 2) a point on the trajectory relative to a fixed point within the plane of the basketball hoop when the basketball is proximate to the basketball hoop.
2. The system of claim 1 wherein the fixed point is a desired target location within the plane of the basketball hoop.
3. The system of claim 2 wherein the desired target location is within a perimeter of the basketball hoop and is a distance from a geometric center of the hoop.
4. The system of claim 1 wherein the system is configured to provide feedback information that allows the human to learn to shoot a shot toward the desired target location.
5. The system of claim 1, wherein the one or more cameras is located on a portable device.
6. The system of claim 1, where portable device is a cell-phone or laptop.
7. The system of claim 6, wherein software is downloaded to the portable device to allow the device to perform as a machine vision system.
8. A training device for use in basketball, said training device comprising:
- a mechanism designed to be coupled to a basketball hoop said mechanism configured to cause shots proximate to a front of the basketball hoop that pass through the hoop without the mechanism installed to be deflected away from the basketball hoop to encourage a user to aim their shots deeper in the basketball rim behind the geometric center of the hoop.
9. The training device of claim 8, wherein the mechanism extends around a portion of the circumference of basketball hoop.
10. The training device of claim 8, wherein the mechanism is configured to allow shots that clear the mechanism to deflect to deflect off a back portion of the basketball rim and pass through the hoop.
11. A system for analyzing a trajectory of a basketball, the device comprising:
- a first device including: a) one or more cameras for recording video frame data used to characterize a trajectory of a basketball shot by a human; b) a logic device designed or configured to i) receive the video frame data, ii) generate trajectory parameters that characterize one or more states of the basketball along its trajectory and iii) generate feedback information using the trajectory parameters; c) one or more feedback output mechanisms for providing the feedback information to the human wherein the feedback information is related to one or more of the following trajectory parameters: 1) an angle of the trajectory relative to a plane of a basketball hoop when the basketball is proximate to the basketball hoop, 2) a point on the trajectory relative to a fixed point within the plane of the basketball hoop when the basketball is proximate to the basketball hoop and d) a first wireless interface
- a second device including a) a video display, b) a storage device for storing data generated by the first device, c) a second wireless interface configured to communicate with the first device via the first wireless interface and to upload data to a remote device separate from the first device.
Filed: Dec 13, 2010
Publication Date: Apr 18, 2013
Patent Grant number: 8617008
Applicant: Pillar Vision, Inc. (Menlo Park, CA)
Inventors: Alan W. Marty (Menlo Park, CA), Thomas A. Edwards (Menlo Park, CA), John Carter (Elkmont, AL)
Application Number: 12/966,301
International Classification: A63B 69/00 (20060101);