mlsanityData to 14 Oct 2026

DM3Net: Dual-Camera Super-Resolution via Domain Modulation and Multi-scale Matching

arXiv 2506.06993 ↗First posted 8 Jun 2025cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Dual-camera super-resolution is highly practical for smartphone photography that primarily super-resolve the wide-angle images using the telephoto image as a reference. In this paper, we propose DM3Net, a novel dual-camera super-resolution network based on Domain Modulation and Multi-scale Matching. To bridge the domain gap between the high-resolution domain and the degraded domain, we learn two compressed global representations from image pairs corresponding to the two domains. To enable reliable transfer of high-frequency structural details from the reference image, we design a multi-scale matching module that conducts patch-level feature matching and retrieval across multiple receptive fields to improve matching accuracy and robustness. Moreover, we also introduce Key Pruning to achieve a significant reduction in memory usage and inference time with little model performance sacrificed. Experimental results on three real-world datasets demonstrate that our DM3Net outperforms the state-of-the-art approaches.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 28 of its 41 references that are in the map. First posted on arXiv 8 Jun 2025.

Builds on

28 of 41 references.

References · 28

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  3. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  4. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  5. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
  6. 2017Squeeze-and-Excitation Networkscited by 4,789
  7. 2016Perceptual Losses for Real-Time Style Transfer and Super-Resolutioncited by 2,694
  8. 2020RAFT: Recurrent All-Pairs Field Transforms for Optical Flowcited by 2,415
  9. 2017Arbitrary Style Transfer in Real-time with Adaptive Instance Normalizationcited by 2,010
  10. 2021SwinIR: Image Restoration Using Swin Transformercited by 1,887
  11. 2017Deformable Convolutional Networkscited by 1,488
  12. 2014Image Super-Resolution Using Deep Convolutional Networkscited by 1,224
16 more
  1. 2021Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Datacited by 970
  2. 2018Deformable ConvNets v2: More Deformable, Better Resultscited by 670
  3. 2020Learning Texture Transformer Network for Image Super-Resolutioncited by 347
  4. 2023DiffIR: Efficient Diffusion Model for Image Restorationcited by 255
  5. 2019Image Super-Resolution by Neural Texture Transfercited by 120
  6. 2022Efficient and Degradation-Adaptive Network for Real-World Image Super-Resolutioncited by 96
  7. 2018CrossNet: An End-to-end Reference-based Super Resolution Network using Cross-scale Warpingcited by 91
  8. 2021MASA-SR: Matching Acceleration and Spatial Adaptation for Reference-Based Image Super-Resolutioncited by 80
  9. 2021Robust Reference-based Super-Resolution via C2-Matchingcited by 69
  10. 2022Reference-based Image Super-Resolution with Deformable Attention Transformercited by 51
  11. 2021Dual-Camera Super-Resolution with Aligned Attention Modulescited by 34
  12. 2022Self-Supervised Learning for Real-World Super-Resolution from Dual Zoomed Observationscited by 20
  13. 2022Coarse-to-Fine Embedded PatchMatch and Multi-Scale Dynamic Aggregation for Reference-based Super-Resolutioncited by 19
  14. 2024Generating Content for HDR Deghosting from Frequency Viewcited by 14
  15. 2023KeDuSR: Real-World Dual-Lens Super-Resolution via Kernel-Free Matchingcited by 4
  16. 2024Efficient Hybrid Zoom using Camera Fusion on Mobile Phonescited by 3

Built on by

Not cited yet.