arxiv:2603.12533

Do You See What I Am Pointing At? Gesture-Based Egocentric Video Question Answering

Published on Mar 13

· Submitted by

Authors:

Abstract

EgoPointVQA presents a dataset and benchmark for gesture-grounded egocentric question answering, along with Hand Intent Tokens (HINT) that encode 3D hand keypoints to improve pointing intent interpretation in multimodal models.

AI-generated summary

Understanding and answering questions based on a user's pointing gesture is essential for next-generation egocentric AI assistants. However, current Multimodal Large Language Models (MLLMs) struggle with such tasks due to the lack of gesture-rich data and their limited ability to infer fine-grained pointing intent from egocentric video. To address this, we introduce EgoPointVQA, a dataset and benchmark for gesture-grounded egocentric question answering, comprising 4000 synthetic and 400 real-world videos across multiple deictic reasoning tasks. Built upon it, we further propose Hand Intent Tokens (HINT), which encodes tokens derived from 3D hand keypoints using an off-the-shelf reconstruction model and interleaves them with the model input to provide explicit spatial and temporal context for interpreting pointing intent. We show that our model outperforms others in different backbones and model sizes. In particular, HINT-14B achieves 68.1% accuracy, on average over 6 tasks, surpassing the state-of-the-art, InternVL3-14B, by 6.6%. To further facilitate the open research, we will release the code, model, and dataset. Project page: https://yuuraa.github.io/papers/choi2026egovqa

View arXiv page View PDF Project page GitHub 1 Add to collection

Community

Yuuraa

Paper submitter about 14 hours ago

Understanding and answering questions based on a user's pointing gesture is essential for next-generation egocentric AI assistants. However, current Multimodal Large Language Models (MLLMs) struggle with such tasks due to the lack of gesture-rich data and their limited ability to perform fine-grained spatial reasoning from egocentric video. To address this, we introduce EgoPointVQA, a dataset and benchmark for gesture-grounded egocentric question answering, comprising 4,000 synthetic and 400 real-world videos across multiple deictic reasoning tasks. Built upon it, we further propose Hand Intent Tokens (HINT), which encode tokens derived from 3D hand keypoints using an off-the-shelf reconstruction model and interleave them with the model input to provide explicit spatial and temporal context for interpreting pointing intent. We show that our model outperforms others across different backbones and model sizes. In particular, HINT-14B achieves 68.1% accuracy on average over 6 tasks, surpassing the state-of-the-art InternVL3-14B by 8.6%.