In one implementation, a method includes receiving pixel events output by an event sensor that correspond to a feature disposed within a field of view of the event sensor. Each respective pixel event is generated in response to a specific pixel within a pixel array of the event sensor detecting a change in light intensity that exceeds a comparator threshold. A characteristic of the feature is determined at a first time based on the pixel events and a previous characteristic of the feature at a second time that precedes the first time. Movement of the feature relative to the event sensor is tracked over time based on the characteristic and the previous characteristic.

In one implementation, a method includes receiving pixel events output by an event sensor that correspond to a feature disposed within a field of view of the event sensor. Each respective pixel event is generated in response to a specific pixel within a pixel array of the event sensor detecting a change in light intensity that exceeds a comparator threshold. A characteristic of the feature is determined at a first time based on the pixel events and a previous characteristic of the feature at a second time that precedes the first time. Movement of the feature relative to the event sensor is tracked over time based on the characteristic and the previous characteristic.

A system and method for performing object detection are presented. The system receives spatial structure information associated with an object which is or has been in a camera field of view of a spatial structure sensing camera. The spatial structure information is generated by the spatial structure sensing camera, and includes depth information for an environment in the camera field of view. The system determines a container pose based on the spatial structure information, wherein the container pose is for describing at least one of an orientation for the container or a depth value for at least a portion of the container. The system further determines an object pose based on the container pose, wherein the object pose is for describing at least one of an orientation for the object or a depth value for at least a portion of the object.

A method of creating a 3D model of a physical object includes adaptively and iteratively generating a number disparity maps from image data representing a plurality of images of the physical object iteratively captured by a plurality of cameras having electrically adjustable focal lengths by varying at least one of the focal lengths of the plurality of cameras and a distance of the physical object from the plurality of cameras during capture of the images until one of the disparity maps is determined to have a least a threshold level of disparity, and converting the one of the disparity maps into the 3D model.

A method of creating a 3D model of a physical object includes adaptively and iteratively generating a number disparity maps from image data representing a plurality of images of the physical object iteratively captured by a plurality of cameras having electrically adjustable focal lengths by varying at least one of the focal lengths of the plurality of cameras and a distance of the physical object from the plurality of cameras during capture of the images until one of the disparity maps is determined to have a least a threshold level of disparity, and converting the one of the disparity maps into the 3D model.

The disclosed subject matter is directed to employing machine learning models configured to predict 3D data from 2D images using deep learning techniques to derive 3D data for the 2D images. In some embodiments, a method is provided that comprises receiving, by a system comprising a processor, a panoramic image, and employing, by the system, a three-dimensional data from two-dimensional data (3D-from-2D) convolutional neural network model to derive three-dimensional data from the panoramic image, wherein the 3D-from-2D convolutional neural network model employs convolutional layers that wrap around the panoramic image as projected on a two-dimensional plane to facilitate deriving the three-dimensional data.

The disclosed subject matter is directed to employing machine learning models configured to predict 3D data from 2D images using deep learning techniques to derive 3D data for the 2D images. In some embodiments, a method is provided that comprises receiving, by a system comprising a processor, a panoramic image, and employing, by the system, a three-dimensional data from two-dimensional data (3D-from-2D) convolutional neural network model to derive three-dimensional data from the panoramic image, wherein the 3D-from-2D convolutional neural network model employs convolutional layers that wrap around the panoramic image as projected on a two-dimensional plane to facilitate deriving the three-dimensional data.

According to an embodiment of the present invention, disclosed is a camera module comprising: an optical output unit for outputting an optical signal to an object; an optical unit for transmitting the optical signal reflected from the object; a sensor for receiving the optical signal transmitted through the optical unit; and a control unit for acquiring the depth map of the object by using the optical signal received by the sensor, wherein the sensor includes an effective area in which a light receiving element is arranged and a non-effective area excluding the effective area, and includes a first row area in which the effective area and the non-effective area are alternately arranged in a row direction, and a second row area in which the effective area and the non-effective area are alternately arranged in the row direction, and in which the effective area is arranged in a column direction at a position not overlapping with the effective area of the first row area, light reaching the effective area of the first row area is controlled by means of first shifting control so as to reach the non-effective area of the first row area or the non-effective area of the second row area, and light reaching the effective area of the second row area is controlled by means of the first shifting control so as to reach the non-effective area of the second row area or the non-effective area of the first row area.

According to an embodiment of the present invention, disclosed is a camera module comprising: an optical output unit for outputting an optical signal to an object; an optical unit for transmitting the optical signal reflected from the object; a sensor for receiving the optical signal transmitted through the optical unit; and a control unit for acquiring the depth map of the object by using the optical signal received by the sensor, wherein the sensor includes an effective area in which a light receiving element is arranged and a non-effective area excluding the effective area, and includes a first row area in which the effective area and the non-effective area are alternately arranged in a row direction, and a second row area in which the effective area and the non-effective area are alternately arranged in the row direction, and in which the effective area is arranged in a column direction at a position not overlapping with the effective area of the first row area, light reaching the effective area of the first row area is controlled by means of first shifting control so as to reach the non-effective area of the first row area or the non-effective area of the second row area, and light reaching the effective area of the second row area is controlled by means of the first shifting control so as to reach the non-effective area of the second row area or the non-effective area of the first row area.

Embodiments of the invention provide a camera array imaging architecture that computes depth maps for objects within a scene captured by the cameras, and use a near-field sub-array of cameras to compute depth to near-field objects and a far-field sub-array of cameras to compute depth to far-field objects. In particular, a baseline distance between cameras in the near-field subarray is less than a baseline distance between cameras in the far-field sub-array in order to increase the accuracy of the depth map. Some embodiments provide an illumination near-IR light source for use in computing depth maps.