FreqKD: Frequency-Decoupled Cross-Modal Knowledge Distillation for Infrared Object Detection

• URL: http://arxiv.org/abs/2606.11572v1
• Date: Wed, 10 Jun 2026 01:55:55 GMT
• Title: FreqKD: Frequency-Decoupled Cross-Modal Knowledge Distillation for Infrared Object Detection
• Abstract summary: Transfer learning from large-scale RGB foundation models to infrared (IR) imagery through knowledge distillation (KD) remains challenging.<n>We propose FreqKD, a frequency-decoupled distillation framework that applies asymmetric supervision adapted to each band's cross-modal consistency.<n>On KAIST multispectral pedestrian detection, FreqKD achieves 64.1 mAP50, improving 2.4 points over the DINOv2 baseline.
• Score: 2.4089107936585235
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Transfer learning from large-scale RGB foundation models to infrared (IR) imagery through knowledge distillation (KD) remains challenging due to fundamental differences in image formation physics. We investigate the spectral structure of the RGB--IR modality gap and observe that feature divergence is not uniform across spatial frequencies: low-frequency components (shape, layout) show greater cross-modal alignment than high-frequency components (texture, fine edges), which reflect modality-specific characteristics. Based on this analysis, we propose FreqKD, a frequency-decoupled distillation framework that applies asymmetric supervision adapted to each band's cross-modal consistency. The method employs strict mean squared error (MSE) on the low-frequency band to preserve shared structural information and a relaxed log-MSE loss (weighted at 0.1) on the high-frequency band to provide edge guidance while tolerating texture differences. Spectral divergence analysis on 500 paired samples shows that high-frequency divergence exceeds low-frequency divergence by a factor of 2.4x on average across all analysed transformer layers. On KAIST multispectral pedestrian detection, FreqKD achieves 64.1 mAP50, improving 2.4 points over the DINOv2 baseline. The learned representation transfers across datasets (FLIR ADAS, +2.1 mAP50), tasks (MFNet segmentation, +1.85 mean intersection-over-union), and architectures (ResNet-50, +1.0 mAP50). Code is available at: https://anonymous.4open.science/r/freq_decoupled_kd-5E5A

Related papers

• S$^3$AM: A Single-Stream SAM with Reliability-Calibrated Frequency Adapter for Multi-modal Salient Object Detection [74.15663758681849]
  Single-stream framework integrates reliability-calibrated frequency adaptation into the adopted SAM backbone for MSOD.<n>Experiments on RGB-D, RGB-T, and RGB-NIR salient object detection benchmarks validate that the proposed framework achieves competitive performance.
  arXiv  Detail & Related papers  (2026-08-18T08:05:00Z)
• RF Spectrogram Anomaly Detection with Quantum Kitchen Sinks: Architecture, Representation, and Hardware Validation [8.643667082966944]
  Quantum Kitchen sinks (QKS) offer a lightweight hybrid quantum feature map suitable for near-term quantum devices.<n>We extend the standard QKS template with multi-depth data re-uploading and ring entanglement, and evaluate the resulting pipeline on controlled RF spectrogram anomaly detection.<n>These results provide a practical, reproducible framework for deploying QKS-based anomaly detection in wireless networks.
  arXiv  Detail & Related papers  (2026-07-15T14:40:41Z)
• PF-Trans: Physics-Embedded Frequency-Aware Transformer for Spectral Reconstruction [9.122692658405116]
  We propose a Physics-embedded Frequency-aware Transformer (PF-Trans) for high-fidelity remote sensing spectral reconstruction.<n>PF-Trans achieves state-of-the-art performance, achieving a Peak Signal-to-Noise Ratio (PSNR) of up to 48.50 dB on the GF-5 Shanghai dataset.
  arXiv  Detail & Related papers  (2026-06-09T03:34:34Z)
• SONAR: Spectral-Contrastive Audio Residuals for Generalizable Deepfake Detection [6.042897432654865]
  Spectral-cONtrastive Audio Residuals (AR) is a frequency-guided framework for deepfake audio detectors.<n>AR disentangles an audio signal into complementary representations.<n> evaluated on the ASVspoof 2021 and in-the-wild benchmarks.
  arXiv  Detail & Related papers  (2025-11-26T12:16:38Z)
• FreqCross: A Multi-Modal Frequency-Spatial Fusion Network for Robust Detection of Stable Diffusion 3.5 Generated Images [4.524282351757178]
  FreqCross is a novel multi-modal fusion network that combines spatial RGB features, frequency domain artifacts, and radial energy distribution patterns.<n>Experiments on a dataset of 10,000 paired real (MS-COCO) and synthetic (Stable Diffusion 3.5) images demonstrate that FreqCross achieves 97.8% accuracy.
  arXiv  Detail & Related papers  (2025-07-01T22:12:35Z)
• Freqformer: Image-Demoiréing Transformer via Efficient Frequency Decomposition [83.40450475728792]
  We present Freqformer, a Transformer-based framework specifically designed for image demoir'eing through targeted frequency separation.<n>Our method performs an effective frequency decomposition that explicitly splits moir'e patterns into high-frequency spatially-localized textures and low-frequency scale-robust color distortions.<n>Experiments on various demoir'eing benchmarks demonstrate that Freqformer achieves state-of-the-art performance with a compact model size.
  arXiv  Detail & Related papers  (2025-05-25T12:23:10Z)
• PAD: Phase-Amplitude Decoupling Fusion for Multi-Modal Land Cover Classification [49.37555541088792]
  Phase-Amplitude Decoupling (PAD) is a frequency-aware framework that separates phase (modality-shared) and amplitude (modality-complementary) components.<n>This work establishes a new paradigm for physics-aware multi-modal fusion in remote sensing.
  arXiv  Detail & Related papers  (2025-04-27T07:21:42Z)
• Bringing RGB and IR Together: Hierarchical Multi-Modal Enhancement for Robust Transmission Line Detection [67.02804741856512]
  We propose a novel Hierarchical Multi-Modal Enhancement Network (HMMEN) that integrates RGB and IR data for robust and accurate TL detection.<n>Our method introduces two key components: (1) a Mutual Multi-Modal Enhanced Block (MMEB), which fuses and enhances hierarchical RGB and IR feature maps in a coarse-to-fine manner, and (2) a Feature Alignment Block (FAB) that corrects misalignments between decoder outputs and IR feature maps by leveraging deformable convolutions.
  arXiv  Detail & Related papers  (2025-01-25T06:21:06Z)
• Deep Fourier-embedded Network for RGB and Thermal Salient Object Detection [15.470610918037243]
  We propose a purely Fourier Transform-based model, namely Deep Fourier-embedded Network (FreqSal) for accurate RGB-T SOD.<n>Specifically, we leverage the efficiency of Fast Fourier Transform with linear complexity to design three key components.<n>Experiments on ten bimodal SOD benchmark datasets demonstrate that FreqSal outperforms twenty-nine existing state-of-the-art bimodal SOD models.
  arXiv  Detail & Related papers  (2024-11-27T14:55:16Z)
• Frequency Domain Modality-invariant Feature Learning for Visible-infrared Person Re-Identification [79.9402521412239]
  We propose a novel Frequency Domain modality-invariant feature learning framework (FDMNet) to reduce modality discrepancy from the frequency domain perspective. Our framework introduces two novel modules, namely the Instance-Adaptive Amplitude Filter (IAF) and the Phrase-Preserving Normalization (PPNorm)
  arXiv  Detail & Related papers  (2024-01-03T17:11:27Z)
• Adaptive Frequency Learning in Two-branch Face Forgery Detection [66.91715092251258]
  We propose Adaptively learn Frequency information in the two-branch Detection framework, dubbed AFD. We liberate our network from the fixed frequency transforms, and achieve better performance with our data- and task-dependent transform layers.
  arXiv  Detail & Related papers  (2022-03-27T14:25:52Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.