Systems and methods for semantically classifying shots in video
First Claim
1. A system for classifying videos based on video content, comprising:
- a processor;
one or more software modules including one or more classifiers;
the one or more software modules operable on the processor to perform the steps of;
receiving a video file, the video file including a plurality of frames;
extracting a subset of frames from the video file;
if one or more frames in the extracted subset of frames comprises a dark frame, discarding the one or more dark frames from the subset;
determining whether each frame in the extracted subset includes content associated with a general content category;
for each frame in the extracted subset that includes content associated with the general content category;
dividing the frame into one or more segments based on image content in each segment;
generating a material classification score vector for each segment, each material classification score vector including one or more material values associated with one or more predefined material content types, wherein each material value represents a probability that the respective segment includes that type of material content;
assigning the material classification score vector for each segment to each respective pixel in the segment;
generating a material arrangement vector for the frame based on the material classification score vectors assigned to each pixel;
generating a scene classification score vector for the frame via one or more scene classifiers, the scene classification score vector including one or more scene classification scores associated with one or more predefined scene categories within the general content category;
determining a representative scene classification score vector for the video file based on the generated scene classification score vectors for each extracted frame in the subset that includes content associated with the general content category; and
associating the video file with the one or more predefined scene categories based on the representative scene classification score vector.
10 Assignments
0 Petitions
Accused Products
Abstract
The present disclosure relates to systems and methods for classifying videos based on video content. For a given video file including a plurality of frames, a subset of frames is extracted for processing. Frames that are too dark, blurry, or otherwise poor classification candidates are discarded from the subset. Generally, material classification scores that describe type of material content likely included in each frame are calculated for the remaining frames in the subset. The material classification scores are used to generate material arrangement vectors that represent the spatial arrangement of material content in each frame. The material arrangement vectors are subsequently classified to generate a scene classification score vector for each frame. The scene classification results are averaged (or otherwise processed) across all frames in the subset to associate the video file with one or more predefined scene categories related to overall types of scene content of the video file.
84 Citations
44 Claims
-
1. A system for classifying videos based on video content, comprising:
-
a processor; one or more software modules including one or more classifiers; the one or more software modules operable on the processor to perform the steps of; receiving a video file, the video file including a plurality of frames;
extracting a subset of frames from the video file;if one or more frames in the extracted subset of frames comprises a dark frame, discarding the one or more dark frames from the subset; determining whether each frame in the extracted subset includes content associated with a general content category; for each frame in the extracted subset that includes content associated with the general content category; dividing the frame into one or more segments based on image content in each segment; generating a material classification score vector for each segment, each material classification score vector including one or more material values associated with one or more predefined material content types, wherein each material value represents a probability that the respective segment includes that type of material content; assigning the material classification score vector for each segment to each respective pixel in the segment; generating a material arrangement vector for the frame based on the material classification score vectors assigned to each pixel; generating a scene classification score vector for the frame via one or more scene classifiers, the scene classification score vector including one or more scene classification scores associated with one or more predefined scene categories within the general content category; determining a representative scene classification score vector for the video file based on the generated scene classification score vectors for each extracted frame in the subset that includes content associated with the general content category; and associating the video file with the one or more predefined scene categories based on the representative scene classification score vector. - View Dependent Claims (2, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 18, 19, 20)
-
-
3. A system for classifying videos based on video content, comprising:
-
a processor; one or more software modules including one or more classifiers; a computer program product that includes a computer-readable medium that is usable by the processor, the medium having stored thereon a sequence of instructions associated with the one or more software modules that when executed by the processor causes the execution of the steps of; receiving a video file, the video file including a plurality of frames; extracting a subset of frames from the video file; if one or more frames in the extracted subset of frames comprises a dark frame, discarding the one or more dark frames from the subset; determining whether each frame in the extracted subset includes content associated with a general content category; for each frame in the extracted subset that includes content associated with the general content category, dividing the frame into one or more segments based on image content in each segment; generating a material classification score vector for each segment, each material classification score vector including one or more material values associated with one or more predefined material content types, wherein each material value represents a probability that the respective segment includes that type of material content; assigning the material classification score vector for each segment to each respective pixel in the segment; generating a material arrangement vector for the frame based on the material classification score vectors assigned to each pixel; classifying the material arrangement vector via the one or more scene classifiers to generate the scene classification score vector for the frame; determining a representative scene classification score vector for the video file based on the generated scene classification score vectors for each extracted frame in the subset that includes content associated with the general content category; and associating the video file with the one or more predefined scene categories based on the representative scene classification score vector. - View Dependent Claims (4, 5, 6)
-
-
17. A system for classifying videos based on video content, comprising:
-
a processor; one or more software modules including one or more classifiers; the one or more software modules operable on the processor to perform the steps of; receiving a video file, the video file including a plurality of frames; extracting a subset of frames from the video file; if one or more frames in the extracted subset of frames comprises a dark frame, discarding the one or more dark frames from the subset; determining whether each frame in the extracted subset includes content associated with a general content category; for each frame in the extracted subset that includes content associated with the general content category; generating a scene classification score vector for the frame via one or more scene classifiers, the scene classification score vector including one or more scene classification scores associated with one or more predefined scene categories within the general content category; determining a representative scene classification score vector for the video file based on the generated scene classification score vectors for each extracted frame in the subset that includes content associated with the general content category; associating the video file with the one or more predefined scene categories based on the representative scene classification score vector; identifying representative scene classification scores in the representative scene classification score vector that exceed a predetermined threshold value; and for each representative scene classification score that exceeds the threshold value, associating the video file with the one or more predefined scene categories associated with the classification scores that exceeded the threshold value.
-
-
21. A method for classifying videos based on video content, comprising the steps of:
-
receiving a video file, the video file including a plurality of frames;
extracting a subset of frames from the video file;if one or more frames in the extracted subset of frames comprises a dark frame, discarding the one or more dark frames from the subset; determining whether each frame in the extracted subset includes content associated with a general content category; for each frame in the extracted subset that includes content associated with the general content category; dividing the frame into one or more segments based on image content in each segment; generating a material classification score vector for each segment, each material classification score vector including one or more material values associated with one or more predefined material content types, wherein each material value represents a probability that the respective segment includes that type of material content; assigning the material classification score vector for each segment to each respective pixel in the segment; generating a material arrangement vector for the frame based on the material classification score vectors assigned to each pixel; classifying the material arrangement vector via the one or more scene classifiers to generate the scene classification score vector for the frame; generating a scene classification score vector for the frame via one or more scene classifiers, the scene classification score vector including one or more scene classification scores associated with one or more predefined scene categories within the general content category; determining a representative scene classification score vector for the video file based on the generated scene classification score vectors for each extracted frame in the subset that includes content associated with the general content category; and labeling the video file according to the one or more predefined scene categories based on the representative scene classification score vector. - View Dependent Claims (22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34)
-
-
35. A method for classifying videos based on video content, comprising:
-
receiving a video file, the video file including a plurality of frames;
extracting a subset of frames from the video file;if one or more frames in the extracted subset of frames comprises a dark frame, discarding the one or more dark frames from the subset; determining whether each frame in the extracted subset includes content associated with a general content category; for each frame in the extracted subset that includes content associated with the general content category; generating a scene classification score vector for the frame via one or more scene classifiers, the scene classification score vector including one or more scene classification scores associated with one or more predefined scene categories within the general content category; determining a representative scene classification score vector for the video file based on the generated scene classification score vectors for each extracted frame in the subset that includes content associated with the general content category; labeling the video file according to the one or more predefined scene categories based on the representative scene classification score vector; identifying representative scene classification scores in the representative scene classification score vector that exceed a predetermined threshold value; and for each representative scene classification score that exceeds the threshold value, associating the video file with the one or more predefined scene categories associated with the classification scores that exceeded the threshold value. - View Dependent Claims (36, 37, 38)
-
-
39. A method for labeling videos based on video content, comprising the steps of:
-
receiving a video file, wherein the video file includes a plurality of frames; extracting a set of frames from the plurality of frames in the video file; for each frame in the extracted set of frames, calculating a probability that the frame includes content associated with a predefined scene category; dividing the frame into one or more segments based on image content in each segment; generating a material classification score vector for each segment, each material classification score vector including one or more material values associated with one or more predefined material content types, wherein each material value represents a probability that the respective segment includes that type of material content; assigning the material classification score vector for each segment to each respective pixel in the segment; generating a material arrangement vector for the frame based on the material classification score vectors assigned to each pixel; determining a representative probability for the set of frames based on the calculated probabilities for each frame; and if the representative probability exceeds a predetermined threshold, associating the scene category with the video file. - View Dependent Claims (40, 41, 42, 43, 44)
-
Specification