ACM MM 2024

EAGLE: Egocentric Aggregated Language-video Engine

Figure: EAGLE: Egocentric Aggregated Language-video Engine

Jing Bi, Yunlong Tang, Luchuan Song, Ali Vosoughi, Nguyen Nguyen, Chenliang Xu

EAGLE-400K, a large-scale instruction-tuning dataset for egocentric video, with a unified benchmark and model for first-person video understanding.

My part. Led end-to-end data gathering, instruction-tuning dataset construction and fine-tuning of vision-language large multimodal models for egocentric video understanding, including EAGLE-400K.

EAGLE-400K, a large-scale instruction-tuning dataset for egocentric video, with a unified benchmark and model for first-person video understanding.

All publications