Abstract: A solution is provided to enhance motion detection and recognition of moving objects associated with various sports by embedding motion sensors into sports instruments such as tennis rackets, badminton rackets, baseball bats and golf clubs, that are swung in a three-dimensional (3D) space. A motion sensor device inserted and locked inside a sports instrument is configured to detect motions associated with movements associated with the sports instrument. The device is shaped to be inserted into a sensor housing attached to a selected location inside the sports instrument. The device includes a motion sensor for detecting motion associated with the movement of the sports instrument, a cover for protecting the motion sensor, and a locking mechanism attached to the sensor housing for locking the motion sensor inside the sensor housing.

Abstract: A system is provided for real-time single-view action recognition for sports videos based on key pose analysis of the sports videos. A training module of the system trains feature models for a sports action distinctively associated with each sports type using a large corpus of training videos. The trained feature models include a player detector for detecting locations of a player in video frames of a training video, a set of key pose identifiers for identifying distinctive poses of a sports action associated with a type of sports, and a meta classifier for determining a likelihood that the sports action has happened in a sports video based on the key poses analysis. Responsive to an input sports video being received for real-time action recognition, a set of trained feature models associated with the sports type of the input video are selected and applied to the input video.

Abstract: A solution is provided to enhance motion detection and recognition of moving objects associated with various sports by intelligently embedding motion sensors into sport instruments such as tennis rackets, badminton rackets, baseball bats and golf clubs, that are swung in a three-dimensional (3D) space. The motion sensors embedded inside the sport instruments are securely locked and are detachable for replacement. A motion sensor inserted and locked inside a sport instrument is configured to detect motions associated with movements associated with the sport instrument. The motion parameters associated with the detected motion are collected through the embedded motion sensor and analyzed by a motion detection and recognition system. Based on the analysis of the motion parameters, various types of sports performance reports and performance improvement recommendations are generated for users of the sport instruments.

Abstract: A solution for composing a multi-angle video from videos captured from different angles is disclosed. A request to compose a multi-angle video is received from a target user, requesting composing a multi-angle video using a video captured by the target user and videos captured by other users nearby the target user. A plurality of videos captured by nearby users are identified based on metadata associated with the captured videos, e.g., geolocation and timestamps of the videos, and presented to the target user. A video sequence including two or more videos from the plurality of videos captured by nearby users is received and the multi-angle video is composed based on the received video sequence.

Abstract: A solution is provided for object tracking in a sports video is disclosed. A determination whether a position of the object was identified in a previous video frame is made. If the position of the object was identified in the previous video frame, a new position of the object is identified in a current video frame based on the identified position of the object in the previous video frame. An expected position of the object is identified based on the identified position of the object in the previous video frame and a trained object classification model. A determination is made whether the new position is consistent with the expected position; if the new position is consistent with the expected position, the new position is stored as the position of the object in the current frame.

Abstract: A solution is provided for recommending sports video content samples to users of a recommendation service with enhanced user experience. The recommendation service generates voting scores for sports video content samples according to a voting method, and selects from among the sports content samples according to the voting scores for the sports video content samples. The voting method is based on in part on motion data for a user's sports motion and the motion data is captured by a motion data device. The sports video content samples, e.g., golf videos, can be classified into multiple classes, e.g., golf swing power related videos, and each class is related to a different aspect of the sports motion.

Abstract: A data acquisition method and device for motion recognition, a motion recognition system and a computer readable storage medium are disclosed. The data acquisition device for motion recognition comprises: an initial motion recognition module adapted to perform an initial recognition with respect to motion data collected by a sensor and provide motion data describing a predefined range around a motion trigger point to a data storage module for storage; a data storage module adapted to store motion data provided from the initial motion recognition module; and a communications module adapted to forward the motion data stored in the data storage module to a motion computing device for motion recognition. The present invention makes an initial selection to the motion data to be transmitted to the motion computing device under the same sampling rate.

Abstract: The invention provides a method of ball game motion recognition, an apparatus for the same, and a motion assisting device. The method comprises: obtaining motion parameters corresponding to each sampling time for a motion; extracting feature points according to predetermined feature point recognition tactics utilizing the motion parameters obtained, in which the feature point recognition tactics comprise recognition tactics of at least three types of the feature points, comprising: power-assisting path early stage corresponding feature point, motion top point corresponding feature point, and ball hitting time corresponding feature point; and recognizing the motion as a predetermined ball game type if the feature points extracted satisfy feature point requirements of the predetermined ball game type.

Abstract: The invention provides a method of confirming motion parameters, an apparatus for the same, and a motion assisting device. The invention obtains and utilizes the motion data of a recognized object sampled at each of the sampling time, comprising the acceleration of the recognized object sampled by a tri-axial accelerometer, the angular velocity of the recognized object sampled by a tri-axial gyroscope, and the angle of the recognized object corresponding to a three-dimensional geomagnetic coordinate system sampled by a tri-axial magnetometer. Feedback calculation is utilized to obtain an actual acceleration at each sampling time from the motion original time to the motion end time, and the actual acceleration is obtained by reducing the acceleration of gravity from the acceleration sampled by a tri-axial accelerometer. The invention reduces the complexity of the system, and the accuracy is less affected by environmental factors, particularly light.