Autogaze Efficient Video Understanding

Reduce video token overhead by 4-100x through autoregressive patch selection, enabling MLLMs to process 1K-frame 4K video efficiently. Uses next-token prediction to identify multi-scale patches that matter. Achieves 19x speedup on vision transformers. Use when processing long, high-resolution videos with MLLMs, have budget constraints on tokens/compute, or need to handle 4K resolution at scale.

adu2021 Updated 7 repo stars

File contents

adu2021/skillxiv commit 2f8ee4855b

Frequently asked questions

npx get-skill add adu2021/autogaze-efficient-video-understanding