notesum.ai

Published at November 22

Efficient Long Video Tokenization via Coordinated-based Patch Reconstruction

cs.CV

cs.AI

Released Date: November 22, 2024

Authors: Huiwon Jang¹, Sihyun Yu¹, Jinwoo Shin¹, Pieter Abbeel², Younggyo Seo²

Aff.: ¹KAIST; ²UC Berkeley

Arxiv: http://arxiv.org/abs/2411.14762v1

Refer to caption