Audio-Visual Transformer Based Crowd Counting
This work addresses crowd estimation, a challenging task in computer vision, by enhancing accuracy through multimodal fusion, though it is incremental as it builds on existing audiovisual approaches.
The paper tackled the problem of crowd counting by proposing an audiovisual multi-task network that integrates visual and audio inputs with a novel run-time modality, achieving up to 33.8% improvement over state-of-the-art methods.
Crowd estimation is a very challenging problem. The most recent study tries to exploit auditory information to aid the visual models, however, the performance is limited due to the lack of an effective approach for feature extraction and integration. The paper proposes a new audiovisual multi-task network to address the critical challenges in crowd counting by effectively utilizing both visual and audio inputs for better modalities association and productive feature extraction. The proposed network introduces the notion of auxiliary and explicit image patch-importance ranking (PIR) and patch-wise crowd estimate (PCE) information to produce a third (run-time) modality. These modalities (audio, visual, run-time) undergo a transformer-inspired cross-modality co-attention mechanism to finally output the crowd estimate. To acquire rich visual features, we propose a multi-branch structure with transformer-style fusion in-between. Extensive experimental evaluations show that the proposed scheme outperforms the state-of-the-art networks under all evaluation settings with up to 33.8% improvement. We also analyze and compare the vision-only variant of our network and empirically demonstrate its superiority over previous approaches.