Interaction-GCN: A Graph Convolutional Network based framework for social interaction recognition in egocentric videos

TL;DR

Interaction-GCN introduces a novel graph convolutional network framework that effectively models social interactions in egocentric videos by combining relational cues, temporal propagation, and motion information, achieving state-of-the-art results.

Contribution

The paper presents a new GCN-based framework for social interaction recognition in egocentric videos, integrating relational cues and temporal modeling.

Findings

Achieves state-of-the-art performance on two datasets.

Validates the effectiveness of relational and temporal modeling.

Demonstrates robustness through ablation studies.

Abstract

In this paper we propose a new framework to categorize social interactions in egocentric videos, we named InteractionGCN. Our method extracts patterns of relational and non-relational cues at the frame level and uses them to build a relational graph from which the interactional context at the frame level is estimated via a Graph Convolutional Network based approach. Then it propagates this context over time, together with first-person motion information, through a Gated Recurrent Unit architecture. Ablation studies and experimental evaluation on two publicly available datasets validate the proposed approach and establish state of the art results.