G-TAD: Sub-Graph Localization for Temporal Action Detection

TL;DR

G-TAD introduces a graph convolutional network approach that models multi-level semantic context as sub-graphs for improved temporal action detection in videos, achieving state-of-the-art results.

Contribution

The paper presents a novel GCN-based framework that adaptively incorporates semantic context into video features and formulates action detection as sub-graph localization.

Findings

Achieves 34.09% mAP on ActivityNet-1.3

Reaches 51.6% IoU@0.5 on THUMOS14

Effective context modeling without extra supervision

Abstract

Temporal action detection is a fundamental yet challenging task in video understanding. Video context is a critical cue to effectively detect actions, but current works mainly focus on temporal context, while neglecting semantic context as well as other important context properties. In this work, we propose a graph convolutional network (GCN) model to adaptively incorporate multi-level semantic context into video features and cast temporal action detection as a sub-graph localization problem. Specifically, we formulate video snippets as graph nodes, snippet-snippet correlations as edges, and actions associated with context as target sub-graphs. With graph convolution as the basic operation, we design a GCN block called GCNeXt, which learns the features of each node by aggregating its context and dynamically updates the edges in the graph. To localize each sub-graph, we also design an SGAlign layer to embed each sub-graph into the Euclidean space. Extensive experiments show that G-TAD is capable of finding effective video context without extra supervision and achieves state-of-the-art performance on two detection benchmarks. On ActivityNet-1.3, it obtains an average mAP of 34.09%; on THUMOS14, it reaches 51.6% at IoU@0.5 when combined with a proposal processing method. G-TAD code is publicly available at https://github.com/frostinassiky/gtad.