DAMSDet: Dynamic Adaptive Multispectral Detection Transformer with Competitive Query Selection and Adaptive Feature Fusion
This work addresses robust full-day object detection for applications like surveillance or autonomous systems, but it is incremental as it builds upon existing transformer-based detection frameworks.
The paper tackles the challenges of highly variable complementary characteristics and modality misalignment in infrared-visible object detection by proposing DAMSDet, which includes a Modality Competitive Query Selection strategy and a Multispectral Deformable Cross-attention module, resulting in significant improvements on four public datasets compared to state-of-the-art methods.
Infrared-visible object detection aims to achieve robust even full-day object detection by fusing the complementary information of infrared and visible images. However, highly dynamically variable complementary characteristics and commonly existing modality misalignment make the fusion of complementary information difficult. In this paper, we propose a Dynamic Adaptive Multispectral Detection Transformer (DAMSDet) to simultaneously address these two challenges. Specifically, we propose a Modality Competitive Query Selection strategy to provide useful prior information. This strategy can dynamically select basic salient modality feature representation for each object. To effectively mine the complementary information and adapt to misalignment situations, we propose a Multispectral Deformable Cross-attention module to adaptively sample and aggregate multi-semantic level features of infrared and visible images for each object. In addition, we further adopt the cascade structure of DETR to better mine complementary information. Experiments on four public datasets of different scenes demonstrate significant improvements compared to other state-of-the-art methods. The code will be released at https://github.com/gjj45/DAMSDet.