feat(embed): add support for video inputs - #185
Draft
Nina-mvH wants to merge 3 commits into
Draft
Conversation
videos frames are embedded as images
Contributor
Author
|
Loading large videos would likely cause OOM errors (like #186 ) |
Prevnets OOM for long videos. No longer warns when empty/one video frame
Nina-mvH
changed the title
30 tasks
Nina-mvH
marked this pull request as ready for review
August 20, 2026 19:28
Nina-mvH
marked this pull request as draft
August 20, 2026 19:29
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Adds support for embedding videos on a frame-by-frame level. Instead of embedding the whole video using specialized models, any multimodal that supports image encoding can be used to embed individual frames from the input video. This appears to be a common practice, though misses temporal feature extraction.
Ideally, it would be nice to also have support for models especially designed for video processing, though it seems the implementation details vary depending on the model. It could be possible though to have some logic similar to that in the
translatetask, and support specific model families.Tested with
sentence-transformers/clip-ViT-B-32onmp4Closes #245