3 papers
cs.CV2025
Bidirectional Action Sequence Learning for Long-term Action Anticipation with Large Language Models
Yuji Sato, Yasunori Ishii, Takayoshi Yamashita
Video-based long-term action anticipation is crucial for early risk detection in areas such as automated driving and robotics. Conventional approaches extract features from past ac…
cs.CV2024
VDMA: Video Question Answering with Dynamically Generated Multi-Agents
Noriyuki Kugo, Tatsuya Ishibashi, Kosuke Ono +1
This technical report provides a detailed description of our approach to the EgoSchema Challenge 2024. The EgoSchema Challenge aims to identify the most appropriate responses to qu…
cs.CV2023
Technical Report for Ego4D Long Term Action Anticipation Challenge 2023
Tatsuya Ishibashi, Kosuke Ono, Noriyuki Kugo +1
In this report, we describe the technical details of our approach for the Ego4D Long-Term Action Anticipation Challenge 2023. The aim of this task is to predict a sequence of futur…