Multiple digital cameras view a large scene, such as a part of a city. Some of the cameras view different parts of that scene, and video feeds from the cameras are processed at a computer to generate a photo-realistic synthetic 3D model of the scene. This enables the scene to be viewed from any viewing angle, including angles that the original, real cameras do not occupy—i.e. as though viewed from a ‘virtual camera’ that can be positioned in any arbitrary position. The 3D model combines both static elements that do not alter in real-time, and also dynamic elements that do alter in real-time or near real-time.

A device and method of dimensioning using digital images and depth data is provided. The device includes a camera and a depth sensing device whose fields of view generally overlap. Segments of shapes belonging to an object identified in a digital image from the camera are identified. Based on respective depth data, from the depth sensing device, associated with each of the segments of the shapes belonging to the object, it is determined whether each of the segments is associated with a same shape belonging to the object. Once all the segments are processed to determine their respective associations with the shapes of the object in the digital image, dimensions of the object are computed based on the respective depth data and the respective associations of the shapes.

An image processing apparatus includes a processor configured to generate vertical direction distribution data which is a distribution of distance values to a vertical direction of a distance image from the distance image having distance values according to distance of a road surface in captured images. There is an extraction of a pixel having the highest frequency value for each of the distance values from pixels in a search area. There is a detection of the road surface based on each pixel extracted.

An apparatus and computer-implemented method for generating a three-dimensional scene and non-transitory tangible computer readable medium thereof are provided. The apparatus comprises a storage unit and a processing unit. The storage unit is stored with an image and a plurality of depth data of the image. The processing unit is configured to generate a depth histogram according to the depth data, generate a region of interest of the image according to the depth histogram, decide a moving path of a virtual camera according to the region of interest, and generate the three-dimensional scene according to the image and the moving path.

An imaging device assembly includes a light source, an imaging device formed with a plurality of imaging elements, and a control device. Each imaging element (10) includes a light receiving portion (21), a first charge storage portion (22) and a second charge storage portion (24), and a first charge transfer control means (23) and a second charge transfer control means (24). Under the control of the control device, the imaging element (10) captures an image of an object on the basis of high-intensity light and stores first image signal charge into the first charge storage portion (22) during a first period, and captures an image of the object on the basis of low-intensity light and stores second image signal charge into the second charge storage portion (24) during a second period. The control device obtains an image signal on the basis of the difference between the first image signal charge and the second image signal charge.

A method and system for segmenting a plurality of images. The method comprises the steps of segmenting the image through a novel clustering technique that is, generating a composite depth map including temporally stable segments of the image as well as segments in subsequent images that have changed. These changes may be determined by determining one or more differences between the temporally stable depth map and segments included in one or more subsequent frames. Thereafter, the portions of the one or more subsequent frames that include segments including changes from their corresponding segments in the temporally stable depth map are processed and are combined with the segments from the temporally stable depth map to compute their associated disparities in one or more subsequent frames. The images may include a pair of stereo images acquired through a stereo camera system at a substantially similar time.

The present invention discloses a multispectral camera external parameter self-calibration algorithm based on edge features, and belongs to the field of image processing and computer vision. Because a visible light camera and an infrared camera belong to different modes, fewer satisfactory point pairs are obtained by directly extracting and matching feature points. In order to solve the problem, the method starts from the edge features, and finds an optimal corresponding position of an infrared image on a visible light image through edge extraction and matching. In this way, a search range is reduced and the number of the satisfactory matched point pairs is increased, thereby more effectively conducting joint self-calibration on the infrared camera and the visible light camera. The operation is simple and results are accurate.

ADAS includes a processing circuit and a memory which stores instructions executable by the processing circuit. The processing circuit executes the instructions to cause the ADAS to receive, from a vehicle that is in motion, a video sequence, generate a position image including at least one object included in the stereo image, generate a second position information associated with the at least one object based on reflected signals received from the vehicle, determine regions each including at least a portion of the at least one object as candidate bounding boxes based on the stereo image and the position image, and selectively adjusting class scores of respective ones of the candidate bounding boxes associated with the at least one object based on whether a respective first position information of the respective ones of the candidate bounding boxes matches the second position information.

To prevent an object that should exist in a virtual viewpoint video image from disappearing. The image processing apparatus generates three-dimensional shape data on a moving object from images based on image capturing from a plurality of viewpoints and outputs the data to the apparatus that generates a virtual viewpoint video image. Then, in a case where it is not possible to generate three-dimensional shape data on an object that behaves as the moving object during a part of the period of the image capturing, three-dimensional shape data on the object generated in the past is output to the apparatus that generates a virtual viewpoint video image.

A processing device generates composite images from a sequence of images. The composite images may be used as frames of video. A foreground/background segmentation is performed at selected frames to extract a plurality of foreground object images depicting a foreground object at different locations as it moves across a scene. The foreground object images are stored to a foreground object list. The foreground object images in the foreground object list are overlaid onto subsequent video frames that follow the respective frames from which they were extracted, thereby generating a composite video.