Fully Rotation-Equivariant Spectral-Spatial Learning for Multispectral Object Detection

Existing multispectral detectors are limited by discrete spec-tral processing, a scale-dependent shift in the relative reliability of spec-tral and spatial cues across pyramid levels, and the lack of explicitrotation-equivariant geometric priors for arbitrarily oriented objects. Totackle these limitations, we propose FressDet, a fully rotation-equivariantspectralx0015spatial learning framework for multispectral object detection,capable of capturing the continuous, ordered nature of spectral struc-ture and enabling reliable spectralx0015spatial fusion across pyramid levelsunder arbitrary in-plane rotations. FressDet integrates three complemen-tary components. Spectral Implicit Warp (SpeIW) enables query-basedspectral resampling via a coordinate-conditioned implicit x001Celd, yieldinga monotone, order-preserving warp. Rotation-Equivariant ConsistencyWeighting (ReCoW) adaptively fuses spectral and spatial branches basedon branch reliability, reinforcing informative cues while suppressing noiseacross pyramid levels. The oriented-aware head exploits group-indexedfeatures to stably predict oriented objects without parameter replication.Taken together, FressDet learns more discriminative and robust spectralx0015spatial representations even under rotational perturbations. By achievingstate-of-the-art performance with 93% fewer parameters on x001Cve publicbenchmarks, FressDet demonstrates its ex001Bectiveness and generalizability.Code is available at https://github.com/Riiluo/FressDet.