FR-DETR: Frequency and Recurrent Feature Refinement for Robust Object Detection under Adverse Weather
This work addresses the problem of robust object detection under adverse weather for autonomous driving and surveillance applications, offering an efficient alternative to enhancer-based approaches.
FR-DETR proposes a detector-centric framework that refines features rather than images for object detection under adverse weather, achieving superior accuracy with significantly lower computational cost compared to enhancer-based methods.
Object detection under adverse weather remains challenging due to severe visual degradations and domain shifts. Existing enhancer-based approaches attempt to improve detection by cascading an enhancer with a detector, but they introduce redundant feature extraction and incur high computational cost with limited accuracy gains when paired with SOTA detectors. We propose FR-DETR, a detector-centric framework that refines features rather than images, focusing enhancement on regions of interest and leveraging frequency-domain cues. Specifically, we design (I) a Frequency Refinement Module that dynamically separates and reweights low- and high-frequency components to improve foreground-background discrimination, and (II) a Recurrent Focus Refinement Module (RFRM) that iteratively refines features using coarse predictions as guidance. Extensive experiments demonstrate that FR-DETR achieves superior detection accuracy under adverse weather while being significantly more computationally efficient than enhancer-based methods. Our implementation is available at https://github.com/ducnt1210/FR-DETR.