Interaction-GCN: A Graph Convolutional Network Based Framework for Social Interaction Recognition in Egocentric Videos

Abstract

In this paper we propose a new framework to categorize social interactions in egocentric videos, we named Interac-tionGCN. Our method extracts patterns of relational and nonrelational 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 (GCN) 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. 1

Visual Event-Based Egocentric Human Action Recognition

“…activity recognition [23,21], social interaction analysis [1,11], video summarisation [8,27], etc. To the best of our knowledge, only conventional cameras have been used for action recognition in the context of egocentric vision, an exception being a very recent work [22], which introduces N-EPIC-Kitchens dataset, an event-based version of EPIC-Kitchens [7], a well-known dataset of egocentric videos captured with conventional wearable cameras.…”