SYSTEM AND METHOD FOR MANAGING DIGITAL VIDEOS USING VIDEO FEATURES

• US 20100182401A1
• Filed: 06/18/2008
• Published: 07/22/2010
• Est. Priority Date: 06/18/2007
• Status: Active Grant

• Current Assignee: Electronics and Telecommunications Research Institute
• Original Assignee: Electronics and Telecommunications Research Institute
• Inventors: Ho, Weon Geun, Lee, Seungiae, Yoo, Wonyoung, Yoon, Young-Suk, Suh, Young Ho, Lee, Sang Kwang, Seo, Yong-Seok, Seo, Jin Soo
• Granted Patent: US 8,477,836 B2
• Time in Patent Office: Days
• US Class Current: 348/42
• CPC Class Codes: H04N 5/91 Television signal processing therefor

Abstract

A method for managing digital videos includes extracting a feature of an input digital video, searching for candidate features having similarity to the extracted feature, and comparing the extracted feature with each of the candidate features to select at least one of the candidate features similar to the extracted feature and provide information on a digital video corresponding to the selected candidate feature.

Claims

1. A method for managing digital videos, comprising:

extracting a feature of an input digital video;

searching for candidate features having similarity to the extracted feature; and

comparing the extracted feature with each of the candidate features to select at least one of the candidate features similar to the extracted feature and provide information on a digital video corresponding to the selected candidate feature.

2. The method of claim 1, wherein extracting the feature of an input digital video includes:

decoding the input digital video to generate digital video frames;

normalizing the digital video frames to a predetermined uniform size; and

extracting temporal and spatial attributes from the video frames with the normalized size, the temporal and spatial attributes being used as the extracted feature.

3. The method of claim 2, wherein normalizing the digital video frames includes resizing width, height, and frame rate of each of the video frames into predetermined reference values.

4. The method of claim 2, wherein extracting temporal and spatial attributes is performed either by extracting temporal attribute information and spatial attributes of the video frames separately, or by extracting both temporal attribute and spatial attribute of the video frames at the same time.

5. The method of claim 2, wherein extracting temporal and spatial attributes includes:

extracting scene interval and a representative frame of each scene from the normalized frames; and

extracting delay length between representative frames or time interval and scene change based on the extracted scene interval and the representative frame of each scene.

6. The method of claim 4, wherein the step of extracting temporal attribute includes the steps of:

dividing each of the normalized frames into image blocks with a predetermined size;

detecting a dominant color from each of the divided image blocks; and

detecting an edge from an edge image of each of the divided image blocks.

7. The method of claim 6, wherein the edge is obtained by projecting the edge image in horizontal and vertical directions, or generating a histogram based on the pattern of the edge image.

8. The method of claim 6, wherein the edge is obtained by calculating a sum of weights relative to strength and direction of the edge image.

9. The method of claim 4, wherein extracting temporal and spatial attributes includes:

storing the normalized frames in a frame buffer by a predetermined frame buffer size;

dividing the stored normalized frames into 3D-based cubics with predetermined width, height and length;

performing 3D transformation on the divided cubics; and

performing linear transformation or non-linear transformation on the transformed cubics to extract the temporal and spatial attributes of the input digital video.

10. The method of claim 9, wherein the normalized frames stored in the frame buffer are moved from a starting position to a predetermined buffer and newly normalized frames are stored in the frame buffer by the predetermined frame buffer size, after processing the stored normalized frames.

11. The method of claim 1, wherein the distance corresponds to Euclidean distance between the feature of the input digital video and each of the features of the candidate digital videos.

12. A system for managing digital videos, comprising:

a feature extracting unit for extracting a feature of an input digital video;

a feature searching unit having a plurality of features for searching for candidate features having similarity to the extracted feature; and

a feature comparing unit for comparing the extracted feature with each of the candidate features to select at least one of the candidate features similar to the extracted feature and provide information on a digital video corresponding to the selected candidate feature.

13. The system of claim 12, wherein the feature extracting unit includes:

a video decoder for decoding the input digital video into RGB video frames;

a spatiotemporal sampler for normalizing the decoded video frames to a uniform size; and

a feature extractor for extracting, as the feature of the input digital video, temporal and spatial attributes of the normalized video frames.

14. The system of claim 13, wherein the feature extractor includes:

a temporal feature extractor for extracting a temporal feature from the normalized frames;

a spatial feature extractor for extracting a spatial feature from the normalized frames; and

a spatiotemporal feature extractor for extracting a spatiotemporal feature from the normalized frames.

15. The system of claim 14, wherein the temporal feature extractor extracts scene interval and a representative frame of each scene from the normalized frames.

16. The system of claim 14, wherein the spatial feature extractor includes:

a block divider for dividing each of the normalized frames into image blocks with a predetermined size;

a dominant color detector for detecting a dominant color from each of the divided image blocks; and

an edge image detector for detecting an edge of each of the divided image blocks from an edge image of each of the divided image blocks.

17. The system of claim 16, wherein the edge image detector obtains the edge by projecting the edge image of each image block in horizontal and vertical directions.

18. The system of claim 16, wherein the edge image detector obtains the edge image by generating a histogram based on the pattern of the edge image of each block, or calculating a sum of weights relative to strength and direction of the edge image.

19. The system of claim 14, wherein the spatiotemporal feature extractor includes:

a frame buffer for temporarily storing the normalized frames by a predetermined buffer size;

a cubic generator for dividing the normalized frames into 3D-based cubics with predetermined width, height and length;

a 3D transformer for performing 3D transformation on the divided cubics; and

a linear/nonlinear transformer for performing linear transformation or nonlinear transformation on the 3D transformed cubics.

20. The system of claim 12, wherein the feature comparing unit includes:

a distance measurer for measuring a distance between the feature of the input digital video and each of the candidate features; and

a candidate verifier for selecting at least one of the candidate features having the distance that falls within a reference range to provide information on a digital video corresponding to the selected candidate feature.