Abstract: Technologies for machine learning with convolutional neural networks (CNNs) and support vector machines (SVMs) include a computing device that may train a deep CNN on a training data set to recognize features of the training data set. The computing device processes the training data set with the CNN after training to extract feature vectors. The computing device trains a multiclass SVM on the feature vectors. The computing device may train a CNN on a training data set to classify the training data set. After training, the computing device may exchange a layer of the CNN with a multiclass SVM. The computing device may use the weights of the exchanged layer to generate the SVM. The computing device may convert the multiclass SVM to a series of binary SVMs. The computing device may generate a reduced set model for each of the binary SVMs. Other embodiments are described and claimed.

Abstract: Technologies for utilizing a bowl-shaped image include a computing device to receive a first fisheye image capturing a first scene and a second fisheye image capturing a second scene overlapping with the first scene at an overlapping region. The computing device generates a combined image of the first fisheye image and the second fisheye image, performs object classification on a region of interest of at least one of the first fisheye image or the second fisheye image to classify an object within the region of interest, and generates a portion of a bowl-shaped image based on the combined image.

Abstract: Methods, apparatuses, and systems may provide for using the motion of a vehicle to estimate the orientation of a camera system of a vehicle relative to the vehicle. Image data may be received from a plurality of cameras positioned on the vehicle, and a first constraint set may be determined for the plurality of cameras based on a plurality of feature points in a ground plane proximate to the vehicle. A second constraint set may be determined based on one or more borders of the vehicle. One or more of the cameras may be automatically calibrated based on the first constraint set and the second constraint set.

Abstract: An apparatus for applying a convolutional neural network (CNN) to a wide-angle camera image is described herein. The apparatus includes a camera, controller, convolution mechanism and a fully connected layer. The camera is to capture a wide-angle image, and the controller is to map the image on a 3D surface. The convolution mechanism is to perform convolution on the 3D surface and the fully connected layer is to classify a plurality of features generated by the convolution mechanism.

Abstract: An apparatus for environment perception using a monitoring system is described herein. The apparatus includes a plurality of sensors, wherein the plurality of sensors is to collect data. The apparatus also includes a controller to estimate motion based on the data and data from a plurality of cameras, wherein the data from the plurality of cameras is processed simultaneously. Additionally, the apparatus includes a matching unit to perform feature matching using the motion estimation and a perception unit to determine a 3D position of points in the environment.

Abstract: A surround view system that can provide a surround view, e.g., a 360° view, from a vehicle by way of cameras positioned at various locations on the vehicle. The cameras can generate image data corresponding to the surround view, and a processing device can process the image data and generate the surround view on a simulated predetermined shape that can be viewed from a display. The simulated predetermined shape can have a flat bottom with a rectangular shape and a rim with a parabolic shape.

Abstract: Techniques are provided for adaptive selection of feature keypoints of an image. An example system may include a contrast statistics calculation circuit configured to generate contrast measurements of regions of the image associated with each of the feature keypoints, and to calculate a mean and variance of the contrast measurements. The system may also include an edge statistics calculation circuit configured to generate ratios of principal curvatures of regions of the image associated with each of the feature keypoints, and to calculate a mean and variance of the ratios of principal curvatures. The system may further include a threshold calculation circuit configured to calculate thresholds based on the mean and variance of the contrast measurements and on the mean and variance of the ratios of principal curvatures; and a keypoint filter circuit configured to filter the set of feature keypoints based on the those thresholds.

Abstract: Techniques are provided for adaptive selection of feature keypoints of an image. An example system may include a contrast statistics calculation circuit configured to generate contrast measurements of regions of the image associated with each of the feature keypoints, and to calculate a mean and variance of the contrast measurements. The system may also include an edge statistics calculation circuit configured to generate ratios of principal curvatures of regions of the image associated with each of the feature keypoints, and to calculate a mean and variance of the ratios of principal curvatures. The system may further include a threshold calculation circuit configured to calculate thresholds based on the mean and variance of the contrast measurements and on the mean and variance of the ratios of principal curvatures; and a keypoint filter circuit configured to filter the set of feature keypoints based on the those thresholds.

Abstract: A system, apparatus, method, and computer readable media for calibration of one or more extrinsic image sensor parameters. A system may calibrate a multi-camera surround view system, for example as may be employed in a vehicle, based on image data comprising one or more ground plane landmarks. The system may determine a ground plane projection of the image data, for example through a Radon transformation. A signal associated with at least one of the one or more ground plane landmarks in the ground plane projection(s) may be determined, for example through application of the projection-slice theorem. The landmark signal may be utilized as a response in an automated calibration loop to optimize one or more extrinsic parameter values associated with the sensors.

Abstract: In order to determine extrinsic parameters of a vehicle vision system or an aspect of a vehicle vision system, road lane markings may be identified in an image captured by a camera and provided to the vehicle vision system. For one or more of the identified road lane markings, a first set of parameters defining an orientation and a position of a line along which a road lane marking extends in an image plane may be determined. Also, a second set of parameters defining an orientation and a position of a line along which a road lane marking extends in a road plane may be determined. A linear transformation that defines a mapping between the first set of parameters and the second set of parameters may be identified. The extrinsic parameters may be established based on the identified linear transformation.

Abstract: Technologies for determining a distance of an object from a vehicle include a computing device to identify an object captured in a fisheye image generated by a fisheye camera of the vehicle. The computing device projects a contour of the identified object on a selected virtual plane that is located outside the vehicle and selected from a predefined set of virtual planes based on a location of the identified object relative to the vehicle. The computing device identifies a bottom of the projected contour on the selected virtual plane and determines an intersection point of an imaginary line with a ground plane coincident with a plane on which the vehicle is positioned. The imaginary line passes through each of the identified bottom of the projected contour and the fisheye camera. The computing device determines a location of the identified object relative to the vehicle based on the determined intersection point and the identified bottom of the projected contour.

Abstract: Techniques are disclosed for improving classification performance in supervised learning. In accordance with some embodiments, a multiclass support vector machine (SVM) having three or more classes may be converted to a plurality of binary problems that then may be reduced via one or more reduced-set methods. The resultant reduced-set (RS) vectors may be combined together in one or more joint lists, along with the original support vectors (SVs) of the different binary classes. Each binary problem may be re-trained using the joint list(s) by applying a reduction factor (RF) parameter to reduce the total quantity of RS vectors. In re-training, different kernel methods can be combined, in accordance with some embodiments. Reduction may be performed until desired classification performance is achieved. The disclosed techniques can be used, for example, to improve classification speed, accuracy, class prioritization, or a combination thereof, in the SVM training phase, in accordance with some embodiments.

Abstract: Methods, apparatuses, and systems may provide for using the motion of a vehicle to estimate the orientation of a camera system of a vehicle relative to the vehicle. Image data may be received from a plurality of cameras positioned on the vehicle, and a first constraint set may be determined for the plurality of cameras based on a plurality of feature points in a ground plane proximate to the vehicle. A second constraint set may be determined based on one or more borders of the vehicle. One or more of the cameras may be automatically calibrated based on the first constraint set and the second constraint set.

Abstract: A system, apparatus, method, and computer readable media for calibration of one or more extrinsic image sensor parameters. A system may calibrate a multi-camera surround view system, for example as may be employed in a vehicle, based on image data comprising one or more ground plane landmarks. The system may determine a ground plane projection of the image data, for example through a Radon transformation. A signal associated with at least one of the one or more ground plane landmarks in the ground plane projection(s) may be determined, for example through application of the projection-slice theorem. The landmark signal may be utilized as a response in an automated calibration loop to optimize one or more extrinsic parameter values associated with the sensors.

Abstract: Technologies for determining a distance of an object from a vehicle include a computing device to identify an object captured in a fisheye image generated by a fisheye camera of the vehicle. The computing device projects a contour of the identified object on a selected virtual plane that is located outside the vehicle and selected from a predefined set of virtual planes based on a location of the identified object relative to the vehicle. The computing device identifies a bottom of the projected contour on the selected virtual plane and determines an intersection point of an imaginary line with a ground plane coincident with a plane on which the vehicle is positioned. The imaginary line passes through each of the identified bottom of the projected contour and the fisheye camera. The computing device determines a location of the identified object relative to the vehicle based on the determined intersection point and the identified bottom of the projected contour.

Abstract: A system for calibrating a vision system of an object. For example, a system for calibrating a surround view system of a vehicle. The system may be configured to capture, via a first and second image sensor, a first and second image of a plurality of objects. Upon capturing such information, the system may be configured to determine a position and orientation of the first and the second image sensor relative to the plurality of objects based on the first and the second captured image. Further, the system may be configured to determine a relative position and orientation between the first and the second image sensor based on a result of the determining a position and orientation of the first and the second image sensor. Also, the system may be configured to determine intrinsic and extrinsic parameters of the first and the second image sensor.

Abstract: Technologies for determining a distance of an object from a vehicle include a computing device to identify an object captured in a fisheye image generated by a fisheye camera of the vehicle. The computing device projects a contour of the identified object on a selected virtual plane that is located outside the vehicle and selected from a predefined set of virtual planes based on a location of the identified object relative to the vehicle. The computing device identifies a bottom of the projected contour on the selected virtual plane and determines an intersection point of an imaginary line with a ground plane coincident with a plane on which the vehicle is positioned. The imaginary line passes through each of the identified bottom of the projected contour and the fisheye camera. The computing device determines a location of the identified object relative to the vehicle based on the determined intersection point and the identified bottom of the projected contour.

Abstract: Technologies for visualizing moving objects on a bowl-shaped image include a computing device to receive a first fisheye image generated by a first fisheye camera and capturing a first scene and a second fisheye image generated by a second fisheye camera and capturing a second scene overlapping with the first scene at an overlapping region. The computing device identifies a moving object in the overlapping region and modifies a projected overlapping image region to visualize the identified moving object on a virtual bowl-shaped projection surface. The projected overlapping image region is projected on the virtual bowl-shaped projection surface and corresponds with the overlapping region captured in the first and second fisheye images.

Abstract: A method for flexibly sintering rare earth permanent magnetic alloy comprises: (1) weighing fine powder of rare earth permanent magnetic alloy, loading the fine powder in moulds, and orientedly compacting the fine powder in a press machine and in inert atmosphere to obtain blanks and loading the blanks into charging boxes; (2) after air between the second conveying vehicle and the first isolating valve of the glove box is replaced with inert gas, opening the two isolating valves connected with each other; wherein after a first rolling wheel transmission in the second conveying vehicle transfers the charging tray into the first chamber of the glove box, the two isolating valves are closed, and the second conveying vehicle leaves; (3) after a first conveying vehicle is coupled with a third isolating valve at an end of the second chamber, locking two matching flanges of the two isolating valves tightly; (4) after the first conveying vehicle is coupled with an isolating valve of a sintering furnace, locking match

Abstract: A vehicle navigation system includes a Kalman filter having a first filter receiving satellite positioning data and a second filter receiving vehicle sensor data. The first filter generates a first state vector estimate and a corresponding first state error covariance matrix. The second filter generates a second state vector estimate and a corresponding second state error covariance matrix. A third filter receives the first and second state vector estimates and the first and second state error covariance matrices, and generates a combined state vector estimate and a corresponding combined state error covariance matrix. A prediction processor generates a predicted state vector estimate and a predicted state error covariance matrix from the combined state vector estimate and the combined state error covariance matrix. The predicted state vector estimate and the predicted state error covariance matrix are provided to the first filter, the second filter, and the third filter.