Skip to content

arjunrs1/AVSpeech-Filtering

Repository files navigation

AVSpeech-Filtering

Scripts for filtering AVSpeech data using a vision and language transformer. The videos were filtered to select samples with high audio-visual correspondence. The filtered data was used for Self-Supervised Visual-Acoustic Matching.

Note: Github often has issues rendering python notebooks, so the analysis notebook can also be viewed here

About

Scripts for filtering AVSpeech data using VQA model

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published