Findings of NAACL 2024

OSCaR: Object State Captioning and State Change Representation

Nguyen Nguyen, Jing Bi, Ali Vosoughi, Yapeng Tian, Pooyan Fazli, Chenliang Xu

Object-state captioning and state-change representation for egocentric video; supported in part by an NIH R01 on accessible video description.

My part. Built the OSCaR codebase and released the dataset and checkpoints on Hugging Face; contributed to the paper.

Paper Code Data Project page

All publications