STEAM: Squeeze and Transform Enhanced Attention Module

Channel and spatial attention mechanisms introduced in earlier work enhance the representational capabilities of deep convolutional neural networks (CNNs) but often increase parameter and computational costs. While recent approaches focus solely on efficient feature context modeling for channel attention, we aim to model both channel and spatial attention comprehensively with minimal parameters and reduced computation. Leveraging the principles of relational modeling in graphs, we introduce a constant-parameter module, $\mathit{\text{STEAM:}}\mathit{\text{Squeeze}}\mathit{\text{and}}\mathit{\text{Transform}}\mathit{\text{Enhanced}}\mathit{\text{Attention}}\mathit{\text{Module}}$, which integrates channel and spatial attention to enhance the representation power of CNNs. To our knowledge, we are the first to propose a graph-based approach for modeling both channel and spatial attention, utilizing concepts from multi-head graph transformers. Additionally, we introduce $\mathit{\text{Output}}\mathit{\text{Guided}}\mathit{\text{Pooling}}$ (OGP), which efficiently captures spatial context to further enhance spatial attention. We extensively evaluate STEAM for large-scale image classification, object detection and instance segmentation on standard benchmark datasets. STEAM achieves a $2\mathrm{\%}$ increase in accuracy over the standard ResNet-50 model with only a meager increase in GFLOPs. Furthermore, STEAM outperforms the leading modules, ECA and GCT, in terms of accuracy while achieving a threefold reduction in GFLOPs. The code will be made available upon acceptance.