mlsanityData to 14 Oct 2026

RAW Image Reconstruction from RGB on Smartphones. NTIRE 2025 Challenge Report

arXiv 2506.01947 ↗First posted 2 Jun 2025eess.IV cs.CVcited by 4

Not on the map. Why?

Abstract (arXiv)

Numerous low-level vision tasks operate in the RAW domain due to its linear properties, bit depth, and sensor designs. Despite this, RAW image datasets are scarce and more expensive to collect than the already large and public sRGB datasets. For this reason, many approaches try to generate realistic RAW images using sensor information and sRGB images. This paper covers the second challenge on RAW Reconstruction from sRGB (Reverse ISP). We aim to recover RAW sensor images from smartphones given the corresponding sRGB images without metadata and, by doing this, ``reverse" the ISP transformation. Over 150 participants joined this NTIRE 2025 challenge and submitted efficient models. The proposed methods and benchmark establish the state-of-the-art for generating realistic RAW data.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 44 of its 69 references that are in the map. First posted on arXiv 2 Jun 2025.

Builds on

44 of 69 references.

References · 44

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  2. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
  3. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 5,045
  4. 2016Perceptual Losses for Real-Time Style Transfer and Super-Resolutioncited by 2,694
  5. 2023Vision Transformers Need Registerscited by 871
  6. 2021Uformer: A General U-Shaped Transformer for Image Restorationcited by 866
  7. 2022Simple Baselines for Image Restorationcited by 708
  8. 2015Cyclical Learning Rates for Training Neural Networkscited by 629
  9. 2019Dynamic Convolution: Attention over Convolution Kernelscited by 356
  10. 2019CondConv: Conditionally Parameterized Convolutions for Efficient Inferencecited by 273
  11. 2018Unprocessing Images for Learned Raw Denoisingcited by 250
  12. 2020CycleISP: Real Image Restoration via Improved Data Synthesiscited by 213
32 more
  1. 2020Replacing Mobile Camera ISP with a Single Deep Learning Modelcited by 108
  2. 2021Invertible Image Signal Processingcited by 94
  3. 2018DeepISP: Towards Learning an End-to-End Image Processing Pipelinecited by 88
  4. 2024Transcending the Limit of Local Window: Advanced Super-Resolution Transformer with Adaptive Token Dictionarycited by 64
  5. 2021Mobile Computational Photography: A Tourcited by 62
  6. 2024The Ninth NTIRE 2024 Efficient Super-Resolution Challenge Reportcited by 57
  7. 2022Omni-Dimensional Dynamic Convolutioncited by 55
  8. 2022Model-Based Image Signal Processors via Learnable Dictionariescited by 43
  9. 2023KBNet: Kernel Basis Network for Image Restorationcited by 33
  10. 2023RawHDR: High Dynamic Range Image Reconstruction from a Single Raw Imagecited by 32
  11. 2024NTIRE 2024 Challenge on Short-form UGC Video Quality Assessment: Methods and Resultscited by 29
  12. 2025The Tenth NTIRE 2025 Efficient Super-Resolution Challenge Reportcited by 26
  13. 2025NTIRE 2025 Challenge on Cross-Domain Few-Shot Object Detection: Methods and Resultscited by 25
  14. 2025NTIRE 2025 Challenge on Day and Night Raindrop Removal for Dual-Focused Images: Methods and Resultscited by 22
  15. 2025The Tenth NTIRE 2025 Image Denoising Challenge Reportcited by 21
  16. 2021Semi-Supervised Raw-to-Raw Mappingcited by 16
  17. 2025NTIRE 2025 Challenge on Event-Based Image Deblurring: Methods and Resultscited by 16
  18. 20231M parameters are enough? A lightweight CNN-based model for medical image segmentationcited by 15
  19. 2025NTIRE 2025 Challenge on Short-form UGC Video Quality Assessment and Enhancement: Methods and Resultscited by 14
  20. 2025NTIRE 2025 Challenge on Efficient Burst HDR and Restoration: Datasets, Methods, and Resultscited by 14
  21. 2022Learned Smartphone ISP on Mobile GPUs with Deep Learning, Mobile AI & AIM 2022 Challenge: Reportcited by 13
  22. 2025NTIRE 2025 Challenge on Image Super-Resolution (x4): Methods and Resultscited by 12
  23. 2025NTIRE 2025 Challenge on UGC Video Enhancement: Methods and Resultscited by 12
  24. 2022Reversed Image Signal Processing and RAW Reconstruction. AIM 2022 Challenge Reportcited by 11
  25. 2023BSRAW: Improving Blind RAW Image Super-Resolutioncited by 11
  26. 2025NTIRE 2025 Challenge on Short-form UGC Video Quality Assessment and Enhancement: KwaiSR Dataset and Studycited by 11
  27. 2023Computational Optics for Mobile Terminals in Mass Productioncited by 10
  28. 2025NTIRE 2025 Challenge on Real-World Face Restoration: Methods and Resultscited by 10
  29. 2025NTIRE 2025 Challenge on Video Quality Enhancement for Video Conferencing: Datasets, Methods and Resultscited by 9
  30. 2025ReRAW: RGB-to-RAW Image Reconstruction via Stratified Sampling for Efficient Object Detection on the Edgecited by 8
  31. 2025NTIRE 2025 challenge on Text to Image Generation Model Quality Assessmentcited by 8
  32. 2024Toward Efficient Deep Blind RAW Image Restorationcited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 1.

4 papers.

  1. 2025NTIRE 2025 Challenge on HR Depth from Images of Specular and Transparent Surfacescited by 3
  2. 2025NTIRE 2025 Image Shadow Removal Challenge Reportcited by 1
  3. 2025NTIRE 2025 Challenge on Low Light Image Enhancement: Methods and Resultscited by 1
  4. 2026SEE Challenge 2026: Event-Guided Brightness Adjustment Across a Broad Illumination Rangecited by 0