MultiFuser: Multimodal Fusion Transformer for Enhanced Driver Action Recognition
This work addresses driver safety and interaction by enhancing action recognition in gloomy, dark car cabins, representing an incremental advance in multimodal fusion methods.
The paper tackles driver action recognition in challenging cabin environments by proposing MultiFuser, a multimodal fusion transformer that integrates features from videos like IR and depth cameras, achieving improved performance as demonstrated on the Drive&Act dataset.
Driver action recognition, aiming to accurately identify drivers' behaviours, is crucial for enhancing driver-vehicle interactions and ensuring driving safety. Unlike general action recognition, drivers' environments are often challenging, being gloomy and dark, and with the development of sensors, various cameras such as IR and depth cameras have emerged for analyzing drivers' behaviors. Therefore, in this paper, we propose a novel multimodal fusion transformer, named MultiFuser, which identifies cross-modal interrelations and interactions among multimodal car cabin videos and adaptively integrates different modalities for improved representations. Specifically, MultiFuser comprises layers of Bi-decomposed Modules to model spatiotemporal features, with a modality synthesizer for multimodal features integration. Each Bi-decomposed Module includes a Modal Expertise ViT block for extracting modality-specific features and a Patch-wise Adaptive Fusion block for efficient cross-modal fusion. Extensive experiments are conducted on Drive&Act dataset and the results demonstrate the efficacy of our proposed approach.