Abstract: In one embodiment, a method of determine and track moving objects in a video, including detecting and extracting regions from accepted frames of a video, matching parts including using the extracted parts of a current frame and matching each part from a previous frame to a region in the current frame, tracking the matched parts to form part tracks, and determining a set of path features for each tracked part path. The determined path features are used to classify each path as that of mover or a static. The method includes clustering the paths of movers, including grouping parts of movers that likely belong to a single object, in order to generate one or more single moving objects and track moving objects. Also a system to carry out the method and a non-transitory computer-readable medium that when executed in a processing system causes carrying out the method.

Abstract: Embodiments of the present invention includes a method of determining MSER on a processing system that includes a parallel processor such as a GPU, the method including method steps carried out in parallel. Embodiments of the present invention also include an apparatus that includes a parallel processor and that is operative to carry out the method, and also logic, e.g., executable instructions configured on a non-transitory computer-readable media operative to carry out the method, and a non-transitory computer-readable medium encoded with such instructions. The method includes assigning to the pixels in an image metapixels of all connected pixels at a level, carrying out watershed in parallel on metapixels to determine catchment basins, and carrying out further steps, including steps executed in parallel to determine the MSER of the image.

Abstract: In one embodiment, a method of operating a processing system includes receiving a reference video and calculating global camera motion parameters for a plurality of the reference video's frames. The method further includes using the camera motion parameters to identify reference video frames corresponding to motion-based events, generating hash values for the identified and neighboring frames in the reference video; and storing the identified frames and hash values. In one version, the method also includes receiving a subject video, calculating global camera motion parameters for the frames of the subject video, using these parameters to identify frames of motion-based events in the subject video, and generating hash values for the identified and neighboring frames in the subject video. The identified frames and hash values of the subject can be compared with those stored for the reference video to evaluate similarity.

Abstract: In particular embodiments, a process relating to the creation of a high-resolution video from a low-resolution video. In a particular embodiment, the process receives as input a sequence of low-resolution video frames. The process first determines a matching score for consecutive frames in the sequence, where the matching score is based on a preliminary global transformation between consecutive frames. From the matching scores, the process determines a set of matching windows and relative-motion estimates and then uses the set and estimates to calculate more thorough global transformations and any residual relative-motion which can be explained using independent object motion and/or optical flow. The process uses the latter global transformation and any independent object motion and/or optical flow to create motion trajectories and to generate high-resolution frames by interpolating low-resolution frames at trajectory-defined points, using regular-to-irregular spatio-temporal interpolation.

Abstract: Methods, apparatuses, and systems directed to video hashing. Video hashing can be used to identify video content in a first video data file with video content of a second video data file. In a particular implementation, video matching or video content identification can be divided into two phases: Feature Extraction and Matching. During a feature extraction phase video material is analyzed and a video fingerprint is generated. The video fingerprint may comprise one video hash or multiple video hashes generated using different video hashing algorithms.

Abstract: In particular embodiments, a process relating to the creation of a high-resolution video from a low-resolution video. In a particular embodiment, the process receives as input a sequence of low-resolution video frames. The process first determines a matching score for consecutive frames in the sequence, where the matching score is based on a preliminary global transformation between consecutive frames. From the matching scores, the process determines a set of matching windows and relative-motion estimates and then uses the set and estimates to calculate more thorough global transformations and any residual relative-motion which can be explained using independent object motion and/or optical flow. The process uses the latter global transformation and any independent object motion and/or optical flow to create motion trajectories and to generate high-resolution frames by interpolating low-resolution frames at trajectory-defined points, using regular-to-irregular spatio-temporal interpolation.