Adobe, Inc.

Sri International

Gajinder Singh

Harpreet Singh Sawhney

Jeffrey Lubin

Manika Puri

Balaji Krishnamurthy

Embodiments including a method and apparatus for indexing a video stream are disclosed. In one embodiment, a method for indexing a video stream comprises accessing a video stream comprising a plurality of frames. For each frame, the method determines salient points computes a cross entropy value for each salient point, and sums the cross entropy values to form a frame information number. A sequence of frame information numbers for the plurality of frames in the video streams forms an index value for the video stream.

Method and apparatus for indexing a video stream

Multibitrate (MBR) live video broadcasting is disclosed in which live video input is copied into a plurality of streams each designated for encoding into a different bitrate. The MBR broadcasting operation is made efficient by performing pre-quantization calculations only the first of the plurality of streams. The results of those calculations are then merely copied to the other streams. Quantization and encoding processes may then be applied to each of the streams to process the streams into their respective, predetermined bitrates.

Enhanced live multibitrate video encoding

A computer implemented method for matching video data to a database containing a plurality of video fingerprints of the type described above, comprising the steps of calculating at least one fingerprint representing at least one query frame from the video data; indexing into the database using the at least one calculated fingerprint to find a set of candidate fingerprints; applying a score to each of the candidate fingerprints; selecting a subset of candidate fingerprints as proposed frames by rank ordering the candidate fingerprints; and attempting to match at least one fingerprint of at least one proposed frame.

Content-based matching of videos using local spatio-temporal fingerprints

A computer implemented method computer implemented method for deriving a fingerprint from video data is disclosed, comprising the steps of receiving a plurality of frames from the video data; selecting at least one key frame from the plurality of frames, the at least one key frame being selected from two consecutive frames of the plurality of frames that exhibiting a maximal cumulative difference in at least one spatial feature of the two consecutive frames; detecting at least one 3D spatio-temporal feature within the at least one key frame; and encoding a spatio-temporal fingerprint based on mean luminance of the at least one 3D spatio-temporal feature. The least one spatial feature can be intensity. The at least one 3D spatio-temporal feature can be at least one Maximally Stable Volume (MSV). Also disclosed is a method for matching video data to a database containing a plurality of video fingerprints of the type described above, comprising the steps of calculating at least one fingerprint representing at least one query frame from the video data; indexing into the database using the at least one calculated fingerprint to find a set of candidate fingerprints; applying a score to each of the candidate fingerprints; selecting a subset of candidate fingerprints as proposed frames by rank ordering the candidate fingerprints; and attempting to match at least one fingerprint of at least one proposed frame based on a comparison of gradient-based descriptors associated with the at least one query frame and the at least one proposed frame.

Growing community of inventors

Noida, India

Gajinder Singh