Claims:WE CLAIM:

1. A method for performing context-based transformation of a video (102), the method comprising:
generating, by a video transformation system (103), a scene descriptor (107) for each of one or more scenes (105) corresponding to the video (102);
generating, by the video transformation system (103), a textual descriptor (109) for each of one or more speech segments related to the one or more scenes (105);
determining, by the video transformation system (103), an audio context descriptor (111) based on semantic analysis of the textual descriptor (109) of each of the one or more speech segments;
correlating, by the video transformation system (103), the audio context descriptor (111) with the scene descriptor (107) for generating a scene context descriptor (113) for each of the one or more scenes (105); and
translating, by the video transformation system (103), each of the one or more scenes (105) using the scene context descriptor (113) for transforming the video (102).

2. The method as claimed in claim 1 comprises eliminating one or more redundant scenes corresponding to the video (102) upon detecting similar scene descriptor (107) for the one or more scenes (105).

3. The method as claimed in claim 2, wherein similarity among the scene descriptor (107) of the one or more scenes (105) is determined by quantifying divergence between the scene descriptor (107) of one or more consecutive scenes.

4. The method as claimed in claim 1, wherein the scene descriptor (107) is generated using one or more parameters comprising objects present in the one or more scenes (105), actions performed by the objects, and attributes of background of the objects in the one or more scenes (105).

5. The method as claimed in claim 1, wherein the scene descriptor (107) comprises labels and description for objects present in the one or more scenes (105).

6. The method as claimed in claim 1, wherein generating the textual descriptor (109) comprises translating each of the one or more speech segments into corresponding text segments using a predetermined conversion technique.

7. A video transformation system (103) for performing context-based transformation of a video (102), the video transformation system (103) comprising:
a processor (203); and
a memory (205), communicatively coupled to the processor (203), wherein the memory (205) stores processor-executable instructions, which on execution, cause the processor (203) to:
generate a scene descriptor (107) for each of one or more scenes (105) corresponding to the video (102);
generate a textual descriptor (109) for each of one or more speech segments related to the one or more scenes (105);
determine an audio context descriptor (111) based on semantic analysis of the textual descriptor (109) of each of the one or more speech segments;
correlate the audio context descriptor (111) with the scene descriptor (107) to generate a scene context descriptor (113) for each of the one or more scenes (105); and
translate each of the one or more scenes (105) using the scene context descriptor (113) to transform the video (102).

8. The video transformation system (103) as claimed in claim 7, wherein the processor (203) eliminates one or more redundant scenes corresponding to the video (102) upon detecting similar scene descriptor (107) for the one or more scenes (105).

9. The video transformation system (103) as claimed in claim 8, wherein the processor (203) quantifies divergence between the scene descriptor (107) of one or more consecutive scenes to determine similarity among the scene descriptor (107) of the one or more scenes (105).

10. The video transformation system (103) as claimed in claim 7, wherein the processor (203) generates the scene descriptor (107) using one or more parameters comprising objects present in the one or more scenes (105), actions performed by the objects, and attributes of background of the objects in the one or more scenes (105).

11. The video transformation system (103) as claimed in claim 7, wherein the scene descriptor (107) comprises labels and description for objects present in the one or more scenes (105).

12. The video transformation system (103) as claimed in claim 7, wherein the processor (203) translates each of the one or more speech segments into corresponding text segments using a predetermined conversion technique to generate the textual descriptor (109).

Dated this 15th day of February 2018

SWETHA S. N
IN/PA-2123
OF K&S PARTNERS
ATTORNEY FOR THE APPLICANT
, Description:TECHNICAL FIELD
The present subject matter is, in general, related to video processing and more particularly, but not exclusively, to a method and system for performing context-based transformation of a video.