Semantic2Graph: Graph-based Multi-modal Feature Fusion for Action Segmentation in Videos

TL;DR

Semantic2Graph introduces a graph-based multi-modal feature fusion method for video action segmentation that models long-term dependencies efficiently, reducing computational costs while improving accuracy over previous models.

Contribution

The paper presents Semantic2Graph, a novel graph-structured approach that effectively captures long-term semantic relationships in videos using multi-modal features and GNNs, outperforming prior methods.

Findings

Semantic2Graph outperforms state-of-the-art methods on GTEA and 50Salads datasets.

Semantic edges improve long-term dependency modeling and accuracy.

The approach reduces computational costs compared to LSTM and Transformer-based models.

Abstract

Video action segmentation have been widely applied in many fields. Most previous studies employed video-based vision models for this purpose. However, they often rely on a large receptive field, LSTM or Transformer methods to capture long-term dependencies within videos, leading to significant computational resource requirements. To address this challenge, graph-based model was proposed. However, previous graph-based models are less accurate. Hence, this study introduces a graph-structured approach named Semantic2Graph, to model long-term dependencies in videos, thereby reducing computational costs and raise the accuracy. We construct a graph structure of video at the frame-level. Temporal edges are utilized to model the temporal relations and action order within videos. Additionally, we have designed positive and negative semantic edges, accompanied by corresponding edge weights, to capture both long-term and short-term semantic relationships in video actions. Node attributes encompass a rich set of multi-modal features extracted from video content, graph structures, and label text, encompassing visual, structural, and semantic cues. To synthesize this multi-modal information effectively, we employ a graph neural network (GNN) model to fuse multi-modal features for node action label classification. Experimental results demonstrate that Semantic2Graph outperforms state-of-the-art methods in terms of performance, particularly on benchmark datasets such as GTEA and 50Salads. Multiple ablation experiments further validate the effectiveness of semantic features in enhancing model performance. Notably, the inclusion of semantic edges in Semantic2Graph allows for the cost-effective capture of long-term dependencies, affirming its utility in addressing the challenges posed by computational resource constraints in video-based vision models.