Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

MIPI 2024 Challenge on Few-shot RAW Image Denoising: Methods and Results

Xin Jin, Chunle Guo, Xiaoming Li, Zongsheng Yue, Chongyi Li, Shangchen Zhou, Ruicheng Feng, Yuekun Dai, Peiqing Yang, Chen Change Loy, Ruoqi Li, Chang Liu, Ziyi Wang, Yao Du, Jingjing Yang, Long Bao, Heng Sun, Xiangyu Kong, Xiaoxia Xing, Jinlong Wu, Yuanyang Xue, Hyunhee Park, Sejun Song, Changho Kim, Jingfan Tan, Wenhan Luo, Zikun Liu, Mingde Qiao, Junjun Jiang, Kui Jiang, Yao Xiao, Chuyang Sun, Jinhui Hu, Weijian Ruan, Yubo Dong, Kai Chen, Hyejeong Jo, Jiahao Qin, Bingjie Han, Pinle Qin, Rui Chai, Pengyuan Wang

TL;DR

The paper reviews the MIPI 2024 Few-shot RAW Image Denoising track, addressing the challenge of denoising RAW images with limited paired data. It details the dataset, evaluation metrics, and three-phase workflow, then surveys seven competing methods that blend synthetic-noise pre-training with careful few-shot fine-tuning and ensembling. Key contributions include calibration-free pipelines, noise-model calibration techniques, and attention-based or transformer-inspired architectures, all aimed at bridging synthetic and real noise domains. The results demonstrate competitive PSNR/SSIM performance and validate the effectiveness of few-shot strategies for practical mobile imaging scenarios, underscoring the track’s impact on rapid, data-efficient denoising development.

Abstract

The increasing demand for computational photography and imaging on mobile platforms has led to the widespread development and integration of advanced image sensors with novel algorithms in camera systems. However, the scarcity of high-quality data for research and the rare opportunity for in-depth exchange of views from industry and academia constrain the development of mobile intelligent photography and imaging (MIPI). Building on the achievements of the previous MIPI Workshops held at ECCV 2022 and CVPR 2023, we introduce our third MIPI challenge including three tracks focusing on novel image sensors and imaging algorithms. In this paper, we summarize and review the Few-shot RAW Image Denoising track on MIPI 2024. In total, 165 participants were successfully registered, and 7 teams submitted results in the final testing phase. The developed solutions in this challenge achieved state-of-the-art erformance on Few-shot RAW Image Denoising. More details of this challenge and the link to the dataset can be found at https://mipichallenge.org/MIPI2024.

MIPI 2024 Challenge on Few-shot RAW Image Denoising: Methods and Results

TL;DR

The paper reviews the MIPI 2024 Few-shot RAW Image Denoising track, addressing the challenge of denoising RAW images with limited paired data. It details the dataset, evaluation metrics, and three-phase workflow, then surveys seven competing methods that blend synthetic-noise pre-training with careful few-shot fine-tuning and ensembling. Key contributions include calibration-free pipelines, noise-model calibration techniques, and attention-based or transformer-inspired architectures, all aimed at bridging synthetic and real noise domains. The results demonstrate competitive PSNR/SSIM performance and validate the effectiveness of few-shot strategies for practical mobile imaging scenarios, underscoring the track’s impact on rapid, data-efficient denoising development.

Abstract

The increasing demand for computational photography and imaging on mobile platforms has led to the widespread development and integration of advanced image sensors with novel algorithms in camera systems. However, the scarcity of high-quality data for research and the rare opportunity for in-depth exchange of views from industry and academia constrain the development of mobile intelligent photography and imaging (MIPI). Building on the achievements of the previous MIPI Workshops held at ECCV 2022 and CVPR 2023, we introduce our third MIPI challenge including three tracks focusing on novel image sensors and imaging algorithms. In this paper, we summarize and review the Few-shot RAW Image Denoising track on MIPI 2024. In total, 165 participants were successfully registered, and 7 teams submitted results in the final testing phase. The developed solutions in this challenge achieved state-of-the-art erformance on Few-shot RAW Image Denoising. More details of this challenge and the link to the dataset can be found at https://mipichallenge.org/MIPI2024.
Paper Structure (16 sections, 4 equations, 7 figures, 1 table)

This paper contains 16 sections, 4 equations, 7 figures, 1 table.

Table of Contents

  1. Introduction
  2. MIPI 2024 Few-shot RAW Image Denoising
  3. Datasets
  4. Evaluation
  5. Challenge Phase
  6. Challenge Results
  7. Methods
  8. MiVideoNR
  9. Samsung
  10. AIIA
  11. MS-Denoimer
  12. MyTurn
  13. HBNU
  14. Erlong Mountain Team
  15. Conclusions
  16. ...and 1 more sections

Figures (7)

  • Figure 1: Illustration of the pre-train stage of team MiVideoNR.
  • Figure 2: Model Framework of Samsung
  • Figure 3: Illustration of the two stages of team AIIA
  • Figure 4: Diagram of MS-Denoimer. (a) The diagram of the Multi-Stage Denoimer. (b) The diagram of the single-stage Denoimer. (c) The diagram of the Spatial Multi-head Self-attention Block (S-MSAB). (d) The diagram of the Channel-wise Multi-head Self-attention Block (C-MSAB). (e) The illustration of the Spatial Multi-head Self-Attention (S-MSA). (f) The illustration of the Channel-wise Multi-head Self-Attention (C-MSA). (g) The illustration of the Gated-DConv Feedforward Network (GDFN).
  • Figure 5: Illustration of the two step finetune strategy of team MyTurn
  • ...and 2 more figures