Towards Online Multi-Modal Social Interaction Understanding

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Li, Xinpeng, Deng, Shijian, Lai, Bolin, Pian, Weiguo, Rehg, James M., Tian, Yapeng
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910090583343104
author Li, Xinpeng
Deng, Shijian
Lai, Bolin
Pian, Weiguo
Rehg, James M.
Tian, Yapeng
author_facet Li, Xinpeng
Deng, Shijian
Lai, Bolin
Pian, Weiguo
Rehg, James M.
Tian, Yapeng
contents In this paper, we introduce a new problem, Online-MMSI, where the model must perform multimodal social interaction understanding (MMSI) using only historical information. Given a recorded video and a multi-party dialogue, the AI assistant is required to immediately identify the speaker's referent, which is critical for real-world human-AI interaction. Without access to future conversational context, both humans and models experience substantial performance degradation when moving from offline to online settings. To tackle the challenges, we propose Online-MMSI-VLM, a novel framework based on multimodal large language models. The core innovations of our approach lie in two components: (1) multi-party conversation forecasting, which predicts upcoming speaker turns and utterances in a coarse-to-fine manner; and (2) socially-aware visual prompting, which highlights salient social cues in each video frame using bounding boxes and body keypoints. Our model achieves state-of-the-art results on three tasks across two datasets, significantly outperforming the baseline and demonstrating the effectiveness of Online-MMSI-VLM. Project page: https://sampson-lee.github.io/online-mmsi-project-page.
format Preprint
id arxiv_https___arxiv_org_abs_2503_19851
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Towards Online Multi-Modal Social Interaction Understanding
Li, Xinpeng
Deng, Shijian
Lai, Bolin
Pian, Weiguo
Rehg, James M.
Tian, Yapeng
Computer Vision and Pattern Recognition
In this paper, we introduce a new problem, Online-MMSI, where the model must perform multimodal social interaction understanding (MMSI) using only historical information. Given a recorded video and a multi-party dialogue, the AI assistant is required to immediately identify the speaker's referent, which is critical for real-world human-AI interaction. Without access to future conversational context, both humans and models experience substantial performance degradation when moving from offline to online settings. To tackle the challenges, we propose Online-MMSI-VLM, a novel framework based on multimodal large language models. The core innovations of our approach lie in two components: (1) multi-party conversation forecasting, which predicts upcoming speaker turns and utterances in a coarse-to-fine manner; and (2) socially-aware visual prompting, which highlights salient social cues in each video frame using bounding boxes and body keypoints. Our model achieves state-of-the-art results on three tasks across two datasets, significantly outperforming the baseline and demonstrating the effectiveness of Online-MMSI-VLM. Project page: https://sampson-lee.github.io/online-mmsi-project-page.
title Towards Online Multi-Modal Social Interaction Understanding
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.19851