S$^3$E-Det: A Strict Spatial-Spectral Equivariant Detector for Multispectral Object Detection

Key points

• Multispectral object detection requires detectors to exploit material-dependent spectral responses while maintaining consistent localization under geometric variations.1
• Existing MOD detectors learn discriminative spectral representations, yet lack a unified mechanism that preserves spectral order, adapts spectral-context fusion across pyramid levels, and regulates geometric responses under rotation.1
• Spectral Order-aware Sampling Calibration (SOSC) imposes local order constraints on input-derived spectral features through first-order variation, while Common-Differential Equivariant Fusion (CDEF) adapts spectral-context aggregation across pyramid levels.1
• Across MODA, HOD3K, and DroneVehicle, S$^3$E-Det combines high detection accuracy with a compact parameter footprint and reduced rotation-induced variation.1

Sources (1)

• [1]S$^3$E-Det: A Strict Spatial-Spectral Equivariant Detector for Multispectral Object Detection
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 20, 09:06 AM

  To address these limitations, we propose S$^3$E-Det, a strict spatial-spectral equivariant detector that couples ordered spectral modeling with transformation-consistent prediction.

  Multispectral object detection requires detectors to exploit material-dependent spectral responses while maintaining consistent localization under geometric variations.

Extractive summary: sentences quoted from the sources.