Pith. sign in

REVIEW 7 cited by

Simple Baselines for Image Restoration

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2204.04676 v4 pith:TH3BWRWM submitted 2022-04-10 cs.CV

Liangyu Chen , Xiaojie Chu , Xiangyu Zhang , Jian Sun This is my paper · ORCID
classification cs.CV
keywords sotaimagebaselinemethodsactivationcomputationalcostsexceeding
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Although there have been significant advances in the field of image restoration recently, the system complexity of the state-of-the-art (SOTA) methods is increasing as well, which may hinder the convenient analysis and comparison of methods. In this paper, we propose a simple baseline that exceeds the SOTA methods and is computationally efficient. To further simplify the baseline, we reveal that the nonlinear activation functions, e.g. Sigmoid, ReLU, GELU, Softmax, etc. are not necessary: they could be replaced by multiplication or removed. Thus, we derive a Nonlinear Activation Free Network, namely NAFNet, from the baseline. SOTA results are achieved on various challenging benchmarks, e.g. 33.69 dB PSNR on GoPro (for image deblurring), exceeding the previous SOTA 0.38 dB with only 8.4% of its computational costs; 40.30 dB PSNR on SIDD (for image denoising), exceeding the previous SOTA 0.28 dB with less than half of its computational costs. The code and the pre-trained models are released at https://github.com/megvii-research/NAFNet.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 7 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. UniRes: Universal Image Restoration for Complex Degradations

    cs.CV 2025-06 reject novelty 7.0 of 10

    A diffusion-based framework that combines task-specific restoration experts during sampling to restore real-world photos with mixed blur, noise, and low resolution.

  2. Splat-based 3D Scene Reconstruction with Extreme Motion-blur

    cs.CV 2026-07 conditional novelty 6.0 of 10

    An RGB-D 3D reconstruction pipeline that estimates camera poses from optical flow and depth, then uses Gaussian Splatting to refine geometry and deblur extreme motion blur without needing COLMAP initialization.

  3. After the Party: Navigating the Mapping From Color to Ambient Lighting

    cs.CV 2025-08 unverdicted novelty 6.0 of 10

    A new paired dataset and Retinex-based network, RLN2, for restoring images captured under multiple colored light sources to ambient-normalized versions.

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

    eess.IV 2025-06 conditional novelty 6.0 of 10

    A new smartphone benchmark shows that efficient models generalize better than large models to unseen camera sensors in RGB-to-RAW reconstruction.

  5. Deblurring in the Wild: A Real-World Image Deblurring Dataset from Smartphone High-Speed Videos

    cs.CV 2025-06 conditional novelty 5.0 of 10

    A new 42,000-pair deblurring benchmark built from iPhone 240 fps videos, on which current state-of-the-art deblurring models underperform.

  6. NTIRE 2025 Image Shadow Removal Challenge Report

    cs.CV 2025-06 conditional novelty 4.0 of 10

    The NTIRE 2025 shadow removal challenge report gives a leaderboard of 17 methods on the WSRD+ dataset and a data alignment upgrade that raises baseline PSNR by about 2 dB.

  7. A Comparative Study of NAFNet Baselines for Image Restoration

    cs.CV 2025-06 conditional novelty 3.0 of 10

    On CIFAR10 with synthetic blur and noise, NAFNet-style simple gating and simplified attention beat conventional alternatives, while GroupNorm matches LayerNorm in PSNR.

Pith tools