</> Code with Papers by Syntology

Datasets › BSD

BSD (Berkeley Segmentation Dataset)

Introduced in A Database of Human Segmented Natural Images and its Application to Evaluating Segmentation Algorithms and Measuring Ecological Statistics1 Jan 2001 archive 2025-07-28

BSD is a dataset used frequently for image denoising and super-resolution. Of the subdatasets, BSD100 is aclassical image dataset having 100 test images proposed by Martin et al.. The dataset is composed of a large variety of images ranging from natural images to object-specific such as plants, people, food etc. BSD100 is the testing set of the Berkeley segmentation dataset BSD300.

Source: A Deep Journey into Super-resolution: A Survey Image Source: https://www.slideshare.net/jbhuang/single-image-super-resolution-from-transformed-selfexemplars-cvpr-2015

HomepageBenchmarksPapers JSON

Benchmarks archive 2025-07-28

All 48 leaderboards whose dataset resolves to this page shown (sort by any header). "First row" is the archive's own first row at snapshot, in the archive's row order; nothing here re-ranks and metric direction is not asserted.

First row (archive order)PaperCode
Image Super-Resolution BSD100 - 4x upscaling DRCT-L PSNR 28.16 DRCT: Saving Image Super-resolution away from... ming053l/drct 71 Compare
Image Super-Resolution BSD100 - 2x upscaling WaveMixSR-V2 PSNR 33.12 WaveMixSR-V2: Enhancing Super-resolution with Higher Efficiency pranavphoenix/WaveMixSR 30 Compare
Image Super-Resolution BSD100 - 3x upscaling Hi-IR-L PSNR 29.67 Hierarchical Information Flow for Generalized Efficient... — 21 Compare
Grayscale Image Denoising BSD68 sigma15 ADL PSNR 32.11 Adversarial Distortion Learning for Medical Image Denoising mogvision/adl 16 Compare
Grayscale Image Denoising BSD68 sigma25 KBNet PSNR 29.54 KBNet: Kernel Basis Network for Image Restoration zhangyi-3/kbnet 16 Compare
Grayscale Image Denoising BSD68 sigma50 ADL PSNR 26.87 Adversarial Distortion Learning for Medical Image Denoising mogvision/adl 15 Compare
Image Super-Resolution BSD100 - 8x upscaling DRLN+ PSNR 25.06 Densely Residual Laplacian Super-Resolution saeed-anwar/DRLN 6 Compare
Density Estimation BSDS300 TAN Log-likelihood 159.8 Transformation Autoregressive Networks — 5 Compare
Color Image Denoising BSD68 sigma15 RIDNet PSNR 34.01 Real Image Denoising with Feature Attention saeed-anwar/RIDNet +2 4 Compare
Color Image Denoising BSD68 sigma25 RIDNet PSNR 31.37 Real Image Denoising with Feature Attention saeed-anwar/RIDNet +2 4 Compare
Blind Super-Resolution BSD100 - 2x upscaling DCLS PSNR 32.04 Deep Constrained Least Squares for Blind Image Super-Resolution megvii-research/dcls-sr +1 3 Compare
Blind Super-Resolution BSD100 - 4x upscaling DCLS PSNR 27.6 Deep Constrained Least Squares for Blind Image Super-Resolution megvii-research/dcls-sr +1 3 Compare
Grayscale Image Denoising BSD200 sigma30 RC-Net PSNR 33.57 Image Restoration Using Deep Regulated Convolutional Networks cswin/RC-Nets 3 Compare
Grayscale Image Denoising BSD200 sigma50 RC-Net PSNR 32.48 Image Restoration Using Deep Regulated Convolutional Networks cswin/RC-Nets 3 Compare
Grayscale Image Denoising BSD200 sigma70 RC-Net PSNR 31.17 Image Restoration Using Deep Regulated Convolutional Networks cswin/RC-Nets 3 Compare
Grayscale Image Denoising BSD68 sigma70 N3Net PSNR 25.14 Neural Nearest Neighbors Networks visinf/n3net +1 3 Compare
Salt-And-Pepper Noise Removal BSD300 Noise Level 30% RTF-Net PSNR 44.56 Residual Transformer Fusion Network for Salt and Pepper... — 3 Compare
Salt-And-Pepper Noise Removal BSD300 Noise Level 50% RTF-Net PSNR 38.03 Residual Transformer Fusion Network for Salt and Pepper... — 3 Compare
Salt-And-Pepper Noise Removal BSD300 Noise Level 70% RTF-Net PSNR 34.96 Residual Transformer Fusion Network for Salt and Pepper... — 3 Compare
Blind Super-Resolution BSD100 - 3x upscaling DCLS PSNR 29.07 Deep Constrained Least Squares for Blind Image Super-Resolution megvii-research/dcls-sr +1 2 Compare
Grayscale Image Denoising BSD200 sigma10 RC-Net PSNR 36.36 Image Restoration Using Deep Regulated Convolutional Networks cswin/RC-Nets 2 Compare
Grayscale Image Denoising BSD68 sigma10 Residual Dense Network + PSNR 34.01 Residual Dense Network for Image Restoration yulunzhang/RDN +2 2 Compare
Grayscale Image Denoising BSD68 sigma30 Residual Dense Network + PSNR 28.58 Residual Dense Network for Image Restoration yulunzhang/RDN +2 2 Compare
Grayscale Image Denoising BSD68 sigma35 FFDNet PSNR 27.73 FFDNet: Toward a Fast and Flexible Solution for CNN... cszn/FFDNet +7 2 Compare
Grayscale Image Denoising BSD68 sigma75 FFDNet PSNR 24.79 FFDNet: Toward a Fast and Flexible Solution for CNN... cszn/FFDNet +7 2 Compare
Image Denoising BSD68 sigma50 MaIR+ PSNR 30.08 MaIR: A Locality- and Continuity-Preserving Mamba for... XLearning-SCU/2025-CVPR-MaIR 2 Compare
Image Super-Resolution BSDS100 - 2x upscaling DBPN-RES-MR64-3 PSNR 32.31 Deep Back-Projection Networks for Single Image Super-resolution alterzero/DBPN-Pytorch +6 2 Compare
Color Image Denoising BSD68 sigma10 Residual Dense Network + PSNR 36.49 Residual Dense Network for Image Restoration yulunzhang/RDN +2 1 Compare
Color Image Denoising BSD68 sigma30 Residual Dense Network + PSNR 30.7 Residual Dense Network for Image Restoration yulunzhang/RDN +2 1 Compare
Color Image Denoising BSD68 sigma35 Deep CNN Denoiser PSNR 29.5 Learning Deep CNN Denoiser Prior for Image Restoration cszn/ircnn +1 1 Compare
Color Image Denoising BSD68 sigma5 Deep CNN Denoiser PSNR 40.36 Learning Deep CNN Denoiser Prior for Image Restoration cszn/ircnn +1 1 Compare
Color Image Denoising BSD68 sigma70 Residual Dense Network + PSNR 26.88 Residual Dense Network for Image Restoration yulunzhang/RDN +2 1 Compare
Color Image Denoising BSD68 sigma75 CSCNet PSNR 26.32 Rethinking the CSC Model for Natural Images drorsimon/CSCNet 1 Compare
Compressive Sensing BSD68 CS=50% AMPA-Net Average PSNR 36.33 AMPA-Net: Optimization-Inspired Attention Neural Network... puallee/AMPA-Net 1 Compare
Compressive Sensing BSDS100 - 2x upscaling AMPA-Net Average PSNR 35.95 AMPA-Net: Optimization-Inspired Attention Neural Network... puallee/AMPA-Net 1 Compare
Grayscale Image Denoising BSD68 sigma20 BUIFD75 (blind) PSNR 29.88 Blind Universal Bayesian Image Denoising with Gaussian... majedelhelou/BUIFD +1 1 Compare
Grayscale Image Denoising BSD68 sigma40 BUIFD75 (blind) PSNR 26.31 Blind Universal Bayesian Image Denoising with Gaussian... majedelhelou/BUIFD +1 1 Compare
Grayscale Image Denoising BSD68 sigma45 BUIFD75 (blind) PSNR 25.67 Blind Universal Bayesian Image Denoising with Gaussian... majedelhelou/BUIFD +1 1 Compare
Grayscale Image Denoising BSD68 sigma5 BUIFD75 (blind) PSNR 37.25 Blind Universal Bayesian Image Denoising with Gaussian... majedelhelou/BUIFD +1 1 Compare
Grayscale Image Denoising BSD68 sigma55 BUIFD75 (blind) PSNR 24.55 Blind Universal Bayesian Image Denoising with Gaussian... majedelhelou/BUIFD +1 1 Compare
Grayscale Image Denoising BSD68 sigma60 BUIFD75 (blind) PSNR 24.05 Blind Universal Bayesian Image Denoising with Gaussian... majedelhelou/BUIFD +1 1 Compare
Grayscale Image Denoising BSD68 sigma65 BUIFD75 (blind) PSNR 23.56 Blind Universal Bayesian Image Denoising with Gaussian... majedelhelou/BUIFD +1 1 Compare
Image Denoising BSD68 sigma30 R3L PSNR 27.67 R3L: Connecting Deep Reinforcement Learning to Recurrent... — 1 Compare
Image Super-Resolution BSD100 - 16x upscaling ABPN PSNR 22.72 Image Super-Resolution via Attention based Back... Holmes-Alan/ABPN 1 Compare
Image Super-Resolution BSD200 - 2x upscaling CSRCNN PSNR 32.92 Cascade Convolutional Neural Network for Image Super-Resolution — 1 Compare
Image Super-Resolution BSDS100 - 4x upscaling DBPN-RES-MR64-3 PSNR 27.82 Deep Back-Projection Networks for Single Image Super-resolution alterzero/DBPN-Pytorch +6 1 Compare
Image Super-Resolution BSDS100 - 8x upscaling DBPN-RES-MR64-3 PSNR 25.05 Deep Back-Projection Networks for Single Image Super-resolution alterzero/DBPN-Pytorch +6 1 Compare
Unified Image Restoration BSD68 sigma25 DA-RCOT Average PSNR (dB) 31.23 Degradation-Aware Residual-Conditioned Optimal Transport... xl-tang3/RCOT +1 1 Compare

Papers archive 2025-07-28

30 shown of 101 papers with a leaderboard row on this dataset's benchmarks, newest first. The archive's own "papers using this dataset" list was never published, so this is the benchmark-backed subset; the archive's count for this dataset is 718. The Syntology column is from Syntology's graph (read 2026-09-24), stated per sample; it is not part of any archive number.

DateSamples run Syntology
Residual Transformer Fusion Network for Salt and Pepper Image Denoising 0 3 13 Feb 2025 not harvested
MaIR: A Locality- and Continuity-Preserving Mamba for Image Restoration 1 2 28 Dec 2024 not harvested
Auto-Encoded Supervision for Perceptual Image Super-Resolution 1 1 28 Nov 2024 ran 2 of 6 samples (4 unverified; 6 pointer-only for licence)
Hierarchical Information Flow for Generalized Efficient Image Restoration 0 3 27 Nov 2024 not harvested
Degradation-Aware Residual-Conditioned Optimal Transport for Unified Image Restoration 2 1 3 Nov 2024 ran 6 of 6 samples (0 unverified; 6 pointer-only for licence)
WaveMixSR-V2: Enhancing Super-resolution with Higher Efficiency 1 2 16 Sep 2024 not harvested
Channel-Partitioned Windowed Attention And Frequency Learning for Single Image Super-Resolution 0 6 23 Jul 2024 not harvested
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution 1 3 8 May 2024 not harvested
DRCT: Saving Image Super-resolution away from Information Bottleneck 1 4 31 Mar 2024 ran 4 of 4 samples (0 unverified; 3 pointer-only for licence)
PaddingFlow: Improving Normalizing Flows with Padding-Dimensional Noise 1 1 13 Mar 2024 not harvested
WaveMixSR: A Resource-efficient Neural Network for Image Super-resolution 1 2 1 Jul 2023 not harvested
Fast and Interpretable Nonlocal Neural Networks for Image Denoising via Group-Sparse Convolutional Dictionary Learning 1 3 2 Jun 2023 not harvested
SwinIA: Self-Supervised Blind-Spot Image Denoising without Convolutions 0 3 9 May 2023 not harvested
Resolution Enhancement Processing on Low Quality Images Using Swin Transformer Based on Interval Dense Connection Strategy 2 3 16 Mar 2023 not harvested
KBNet: Kernel Basis Network for Image Restoration 1 3 6 Mar 2023 ran 6 of 9 samples (3 unverified)
Perception-Oriented Single Image Super-Resolution using Optimal Objective Estimation 1 1 24 Nov 2022 not harvested
SwinFIR: Revisiting the SwinIR with Fast Fourier Convolution and Improved Training for Image Super-Resolution 2 6 24 Aug 2022 not harvested
Activating More Pixels in Image Super-Resolution Transformer 2 6 9 May 2022 not harvested
Adversarial Distortion Learning for Medical Image Denoising 1 3 29 Apr 2022 not harvested
Deep Constrained Least Squares for Blind Image Super-Resolution 2 3 15 Feb 2022 ran 0 of 6 samples (6 unverified)
Flexible Style Image Super-Resolution using Conditional Objective 1 4 13 Jan 2022 not harvested
Restormer: Efficient Transformer for High-Resolution Image Restoration 13 1 18 Nov 2021 ran 4 of 4 samples (0 unverified; 4 pointer-only for licence)
Local Texture Estimator for Implicit Representation Function 1 3 17 Nov 2021 ran 6 of 9 samples (3 unverified)
SwinIR: Image Restoration Using Swin Transformer 9 1 23 Aug 2021 ran 30 of 45 samples (15 unverified; 5 pointer-only for licence)
R3L: Connecting Deep Reinforcement Learning to Recurrent Neural Networks for Image Denoising via Residual Recovery 0 1 12 Jul 2021 not harvested
One-to-many Approach for Improving Super-Resolution 1 1 19 Jun 2021 not harvested
CDLNet: Robust and Interpretable Denoising Through Deep Convolutional Dictionary Learning 1 3 5 Mar 2021 not harvested
Pre-Trained Image Processing Transformer 6 1 1 Dec 2020 not harvested
AMPA-Net: Optimization-Inspired Attention Neural Network for Deep Compressed Sensing 1 2 14 Oct 2020 not harvested
Unfolding the Alternating Optimization for Blind Super Resolution 1 2 6 Oct 2020 ran 5 of 5 samples (0 unverified; 5 pointer-only for licence)

The full list of 101 is in the JSON twin.

Dataset loaders archive 2025-07-28

No loader listed in the archive.

Tasks archive 2025-07-28

Image Super-ResolutionColor Image DenoisingGrayscale Image DenoisingImage DenoisingBlind Super-ResolutionDensity EstimationUnified Image RestorationSalt-And-Pepper Noise RemovalCompressive Sensing

License archive 2025-07-28

Custom (research-only, non-commercial, attribution)

Modalities archive 2025-07-28

Images

Languages archive 2025-07-28

No language tagged.

Variants archive 2025-07-28

  • BSD68 sigma65
  • BSD68 sigma60
  • BSD68 sigma55
  • BSD68 sigma45
  • BSD68 sigma40
  • BSD68 sigma20
  • BSD68 CS=50%
  • BSDS300
  • BSDS100 - 8x upscaling
  • BSDS100 - 4x upscaling
  • BSDS100 - 2x upscaling
  • BSD68 sigma75
  • BSD68 sigma70
  • BSD68 sigma50
  • BSD68 sigma5
  • BSD68 sigma35
  • BSD68 sigma30
  • BSD68 sigma25
  • BSD68 sigma15
  • BSD68 sigma10
  • BSD200 sigma70
  • BSD200 sigma50
  • BSD200 sigma30
  • BSD200 sigma10
  • BSD200 - 2x upscaling
  • BSD
  • BSD300 sigma70
  • BSD300 sigma50
  • BSD300 sigma30
  • BSD300 Noise Level 70%
  • BSD300 Noise Level 50%
  • BSD300 Noise Level 30%
  • BSD100 - 8x upscaling
  • BSD100 - 4x upscaling
  • BSD100 - 3x upscaling
  • BSD100 - 2x upscaling
  • BSD100 - 16x upscaling

37 variant names, as the archive lists them.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections