Egocentric Diarization

[jfResearchEng]

🚀 Feature

EGO4D is the world's largest egocentric (first person) video ML dataset and benchmark suite, with 3,600 hrs (and counting) of densely narrated video and a wide range of annotations across five new benchmark tasks. It covers hundreds of scenarios (household, outdoor, workplace, leisure, etc.) of daily life activity captured in-the-wild by 926 unique camera wearers from 74 worldwide locations and 9 different countries. Portions of the video are accompanied by audio, 3D meshes of the environment, eye gaze, stereo, and/or synchronized videos from multiple egocentric cameras at the same event.

This project focuses on the following use cases:
Audio-visual speaker diarization: Given an egocentric video clip, identify which person spoke and when they spoke.
Speech transcription: Given an egocentric video clip, transcribe the speech of each person.
Use Ego4D video clip

Additional context

Example application can be found here
The code should be added at folder is https://github.com/facebookresearch/labgraph/tree/main/extensions/labgraph_diarization
Create setup.py and README.md, where example can be found at: https://github.com/facebookresearch/labgraph/tree/main/extensions/labgraph_viz
Add github action support, reference: https://github.com/facebookresearch/labgraph/actions/workflows/main.yml
Add proper license header.