Unleashing the Power of CNN and Transformer for Balanced RGB-Event Video Recognition

TL;DR

This paper introduces TSCFormer, a lightweight CNN-Transformer framework that effectively combines local and global features for RGB-Event video recognition, achieving a good balance between accuracy and model complexity.

Contribution

The paper proposes a novel CNN-Transformer model, TSCFormer, that fuses RGB and Event data using global tokens and interactive modules, improving recognition performance while maintaining simplicity.

Findings

Validated on large-scale RGB-Event datasets PokerEvent and HARDVS

Achieved superior accuracy with fewer parameters compared to existing methods

Demonstrated effective global-local feature fusion for video recognition

Abstract

Pattern recognition based on RGB-Event data is a newly arising research topic and previous works usually learn their features using CNN or Transformer. As we know, CNN captures the local features well and the cascaded self-attention mechanisms are good at extracting the long-range global relations. It is intuitive to combine them for high-performance RGB-Event based video recognition, however, existing works fail to achieve a good balance between the accuracy and model parameters, as shown in Fig.~\ref{firstimage}. In this work, we propose a novel RGB-Event based recognition framework termed TSCFormer, which is a relatively lightweight CNN-Transformer model. Specifically, we mainly adopt the CNN as the backbone network to first encode both RGB and Event data. Meanwhile, we initialize global tokens as the input and fuse them with RGB and Event features using the BridgeFormer module. It captures the global long-range relations well between both modalities and maintains the simplicity of the whole model architecture at the same time. The enhanced features will be projected and fused into the RGB and Event CNN blocks, respectively, in an interactive manner using F2E and F2V modules. Similar operations are conducted for other CNN blocks to achieve adaptive fusion and local-global feature enhancement under different resolutions. Finally, we concatenate these three features and feed them into the classification head for pattern recognition. Extensive experiments on two large-scale RGB-Event benchmark datasets (PokerEvent and HARDVS) fully validated the effectiveness of our proposed TSCFormer. The source code and pre-trained models will be released at https://github.com/Event-AHU/TSCFormer.