Video based question and answer
Disclosed are systems and methods that convert digital video data, such as two-dimensional digital video data, into a natural language text description describing the subject matter represented in the video. For example, the disclosed implementations may process video data in real-time, near real-time, or after the video data is created and generate a text-based video narrative describing the subject matter of the video. In addition, the disclosed implementations may also support a question and answer session in which a user may submit queries about the subject matter of one or more videos and the disclosed implementations will present natural language responses based on the subject matter of the video and any corresponding context.
1. A computing system, comprising:
one or more processors; and
a memory storing program instructions that, when executed by the one or more processors, cause the one or more processors to at least:
receive a query about a subject matter of a video;
generate, based at least in part on the query, a query embedding;
determine, based at least in part on a distance in a vector space between the query embedding and feature embeddings generated for each of a plurality of segments, the plurality of segments of the video that represent subject matter that is responsive to the query;
generate, based at least in part on the feature embeddings generated for each of the plurality of segments, a natural language response that describes the subject matter represented in the plurality of segments and that is responsive to the query; and
present the natural language response as a response to the query.
2. The computing system of claim 1 :
wherein the program instructions that, when executed by the one or more processors, further cause the one or more processors to at least determine an activity of a plurality of activities that is occurring in the subject matter of the video; and
wherein the program instructions that cause the one or more processors to generate the natural language response further include program instructions that, when executed by the one or more processors, further cause the one or more processors to at least, generate the natural language response based at least in part on the activity and the feature embeddings.
3. The computing system of claim 1 :
wherein the program instructions that, when executed by the one or more processors, further cause the one or more processors to at least determine, based at least in part on the query embedding, a second plurality of segments of a second video that represent subject matter that is responsive to the query; and
wherein the program instructions that cause the one or more processors to generate the natural language response, further include program instructions that, when executed by the one or more processors, further cause the one or more processors to at least determine based at least in part on the feature embeddings generated for each of the plurality of segments of the video and feature embeddings generated for each of the second plurality of segments, the natural language response.
4. The computing system of claim 1 , wherein the program instructions that cause the one or more processors to present the natural language response, further include program instructions that, when executed by the one or more processors, further cause the one or more processors to at least:
present the natural language response at least one of audibly, visually, or haptically.
5. The computing system of claim 1 , wherein the program instructions that cause the one or more processors to generate the natural language response, further include program instructions that, when executed by the one or more processors, further cause the one or more processors to at least:
generate, based at least in part on the feature embeddings generated for each of the plurality of segments and a context determined for the query, the natural language response.
6. The computing system of claim 5 , wherein the context is determined based at least in part on one or more of the query, a query and answer session in which the query is presented, the video, or a user.
7. The computing system of claim 1 :
wherein the program instructions that, when executed by the one or more processors, further cause the one or more processors to at least, determine a plurality of keywords for the query; and
wherein the program instructions that determine the plurality of segments further include program instructions that, when executed by the one or more processors, further cause the one or more processors to at least determine, based at least in part on the query embedding and the plurality of keywords, the plurality of segments.
8. The computing system of claim 1 , wherein the program instructions that, when executed by the one or more processors, further cause the one or more processors to at least:
process each segment of the plurality of segments of the video to generate, for each of the plurality of segments, the feature embedding that corresponds to a subject matter of the embedding.
9. The computing system of claim 8 , wherein the program instructions that cause the one or more processors to process each segment of the plurality of segments, further include program instructions that, when executed by the one or more processors, further cause the one or more processors to at least:
process each segment of the plurality of segments with a temporal semantic attention transformer to generate a plurality of temporal embeddings, wherein the plurality of temporal embeddings are included in the feature embedding for the segment; and
process each segment with an activity semantic attention transformer to generate a plurality of activity embeddings, wherein the plurality of activity embeddings are included in the feature embedding for the segment.
10. A method, comprising:
receiving a query about a subject matter of a video;
generating, based at least in part on the query, a query embedding;
determining, based at least in part on a distance in a vector space between the query embedding and feature embeddings generated for each of a plurality of segments, the plurality of segments of the video that represent subject matter that is responsive to the query;
generating, based at least in part on the feature embeddings generated for each of the plurality of segments, a natural language response that describes the subject matter represented in the plurality of segments and that is responsive to the query; and
presenting the natural language response as a response to the query.
11. The method of claim 10 , further comprising:
determining an activity of a plurality of activities that is occurring in the subject matter of the video; and
wherein generating the natural language response further includes:
generating the natural language response based at least in part on the activity and the feature embeddings.
12. The method of claim 10 , further comprising:
determining, based at least in part on the query embedding, a second plurality of segments of a second video that represent subject matter that is responsive to the query; and
wherein generating the natural language response, further includes:
determining based at least in part on the feature embeddings generated for each of the plurality of segments of the video and feature embeddings generated for each of the second plurality of segments, the natural language response.
13. The method of claim 10 , further comprising:
determining a plurality of keywords for the query; and
wherein determining the plurality of segments further includes determining, based at least in part on the query embedding and the plurality of keywords, the plurality of segments.
14. The method of claim 10 , further comprising:
processing each segment of the plurality of segments of the video to generate, for each of the plurality of segments, the feature embedding that corresponds to a subject matter of the embedding.
15. A non-transitory computer-readable storage medium storing instructions, the instructions, when executed by one or more processors, cause the one or more processors to at least:
receive a query about a subject matter of a video;
generate, based at least in part on the query, a query embedding;
determine, based at least in part on a distance in a vector space between the query embedding and feature embeddings generated for each of a plurality of segments, the plurality of segments of the video that represent subject matter that is responsive to the query;
generate, based at least in part on the feature embeddings generated for each of the plurality of segments, a natural language response that describes the subject matter represented in the plurality of segments and that is responsive to the query; and
present the natural language response as a response to the query.
16. The non-transitory computer-readable storage medium of claim 15 , wherein the instruction that cause the one or more processors to present the natural language response, further cause the one or more processors to at least:
present the natural language response at least one of audibly, visually, or haptically.
17. The non-transitory computer-readable storage medium of claim 15 , wherein the instructions that cause the one or more processors to generate the natural language response, further cause the one or more processors to at least:
generate, based at least in part on the feature embeddings generated for each of the plurality of segments and a context determined for the query, the natural language response.
18. The non-transitory computer-readable storage medium of claim 17 , wherein the context is determined based at least in part on one or more of the query, a query and answer session in which the query is presented, the video, or a user.
19. The non-transitory computer-readable storage medium of claim 15 , wherein the instructions, when executed by the one or more processors, further cause the one or more processors to at least:
process each segment of the plurality of segments of the video to generate, for each of the plurality of segments, the feature embedding that corresponds to a subject matter of the embedding.
20. The non-transitory computer-readable storage medium of claim 19 , wherein the instructions that cause the one or more processors to process each segment of the plurality of segments, further cause the one or more processors to at least:
process each segment of the plurality of segments with a temporal semantic attention transformer to generate a plurality of temporal embeddings, wherein the plurality of temporal embeddings are included in the feature embedding for the segment; and
process each segment with an activity semantic attention transformer to generate a plurality of activity embeddings, wherein the plurality of activity embeddings are included in the feature embedding for the segment.