Vision Backbone Enhancement via Multi-Stage Cross-Scale Attention
This work addresses a specific bottleneck in vision architectures for researchers and practitioners, offering an incremental improvement through a novel attention module.
The paper tackled the limitation of CNNs and ViTs in lacking interactions between feature maps from different stages and scales by proposing the Multi-Stage Cross-Scale Attention (MSCSA) module, which achieved a significant performance boost with modest additional computational costs in experiments on downstream tasks.
Convolutional neural networks (CNNs) and vision transformers (ViTs) have achieved remarkable success in various vision tasks. However, many architectures do not consider interactions between feature maps from different stages and scales, which may limit their performance. In this work, we propose a simple add-on attention module to overcome these limitations via multi-stage and cross-scale interactions. Specifically, the proposed Multi-Stage Cross-Scale Attention (MSCSA) module takes feature maps from different stages to enable multi-stage interactions and achieves cross-scale interactions by computing self-attention at different scales based on the multi-stage feature maps. Our experiments on several downstream tasks show that MSCSA provides a significant performance boost with modest additional FLOPs and runtime.