paper-with-me

Image Super-Resolution 벤치마크

Image Super-Resolution on Set5 - 2x upscaling

164개 결과 · ⬇ CSV · JSON

PSNR

35.05 36.07 37.09 38.12 39.14 2015-11 2026-11 DRCN [[Kim et al.2016b]] — 37.63 (2015-11-14) VDSR [[Kim et al.2016a]] — 37.53 (2015-11-14) DRCN [[Kim et al.2016b]] — 37.63 (2015-11-14) VDSR [[Kim et al.2016a]] — 37.53 (2015-11-14) DRCN [[Kim et al.2016b]] — 37.63 (2015-11-14) VDSR [[Kim et al.2016a]] — 37.53 (2015-11-14) DRCN [[Kim et al.2016b]] — 37.63 (2015-11-14) VDSR [[Kim et al.2016a]] — 37.53 (2015-11-14) RED30 — 37.66 (2016-06-29) RED30 — 37.66 (2016-06-29) RED30 — 37.66 (2016-06-29) RED30 — 37.66 (2016-06-29) DnCNN-3 — 37.58 (2016-08-13) DnCNN-3 — 37.58 (2016-08-13) DnCNN-3 — 37.58 (2016-08-13) DnCNN-3 — 37.58 (2016-08-13) Deep CNN Denoiser — 35.05 (2017-04-11) Deep CNN Denoiser — 35.05 (2017-04-11) Deep CNN Denoiser — 35.05 (2017-04-11) Deep CNN Denoiser — 35.05 (2017-04-11) c-DCSCN — 37.13 (2017-07-18) c-DCSCN — 37.13 (2017-07-18) c-DCSCN — 37.13 (2017-07-18) c-DCSCN — 37.13 (2017-07-18) CARN [[Ahn et al.2018]] — 37.76 (2018-03-23) CARN [[Ahn et al.2018]] — 37.76 (2018-03-23) CARN [[Ahn et al.2018]] — 37.76 (2018-03-23) CARN [[Ahn et al.2018]] — 37.76 (2018-03-23) MWCNN — 37.91 (2018-05-18) MWCNN — 37.91 (2018-05-18) MWCNN — 37.91 (2018-05-18) MWCNN — 37.91 (2018-05-18) N3Net — 37.57 (2018-10-30) N3Net — 37.57 (2018-10-30) N3Net — 37.57 (2018-10-30) N3Net — 37.57 (2018-10-30) FALSR-A — 37.82 (2019-01-22) FALSR-A — 37.82 (2019-01-22) FALSR-A — 37.82 (2019-01-22) FALSR-A — 37.82 (2019-01-22) LFFN-S — 37.66 (2019-02-15) LFFN-S — 37.66 (2019-02-15) LFFN-S — 37.66 (2019-02-15) LFFN-S — 37.66 (2019-02-15) SRFBN — 38.11 (2019-03-23) SRFBN — 38.11 (2019-03-23) SRFBN — 38.11 (2019-03-23) SRFBN — 38.11 (2019-03-23) DBPN-RES-MR64-3 — 38.08 (2019-04-04) DBPN-RES-MR64-3 — 38.08 (2019-04-04) DBPN-RES-MR64-3 — 38.08 (2019-04-04) DBPN-RES-MR64-3 — 38.08 (2019-04-04) HBPN — 38.13 (2019-06-17) HBPN — 38.13 (2019-06-17) HBPN — 38.13 (2019-06-17) HBPN — 38.13 (2019-06-17) DRLN+ — 38.34 (2019-06-28) DRLN+ — 38.34 (2019-06-28) DRLN+ — 38.34 (2019-06-28) DRLN+ — 38.34 (2019-06-28) IMDN — 38.0 (2019-09-26) IMDN — 38.0 (2019-09-26) IMDN — 38.0 (2019-09-26) IMDN — 38.0 (2019-09-26) RC-Net — 37.42 (2019-10-19) RC-Net — 37.42 (2019-10-19) RC-Net — 37.42 (2019-10-19) RC-Net — 37.42 (2019-10-19) CSNLN — 38.28 (2020-06-02) CSNLN — 38.28 (2020-06-02) CSNLN — 38.28 (2020-06-02) CSNLN — 38.28 (2020-06-02) PMRN+ — 38.22 (2020-07-19) PMRN+ — 38.22 (2020-07-19) PMRN+ — 38.22 (2020-07-19) PMRN+ — 38.22 (2020-07-19) SPBP-L+ — 38.05 (2020-08-03) SPBP-L+ — 38.05 (2020-08-03) SPBP-L+ — 38.05 (2020-08-03) SPBP-L+ — 38.05 (2020-08-03) HAN+ — 38.33 (2020-08-20) HAN+ — 38.33 (2020-08-20) HAN+ — 38.33 (2020-08-20) HAN+ — 38.33 (2020-08-20) CSRCNN — 37.45 (2020-08-24) CSRCNN — 37.45 (2020-08-24) CSRCNN — 37.45 (2020-08-24) CSRCNN — 37.45 (2020-08-24) LTE — 38.33 (2021-11-17) LTE — 38.33 (2021-11-17) LTE — 38.33 (2021-11-17) LTE — 38.33 (2021-11-17) EDT-B — 38.63 (2021-12-19) EDT-B — 38.63 (2021-12-19) EDT-B — 38.63 (2021-12-19) EDT-B — 38.63 (2021-12-19) HAT-L — 38.91 (2022-05-09) HAT — 38.73 (2022-05-09) HAT-L — 38.91 (2022-05-09) HAT — 38.73 (2022-05-09) HAT-L — 38.91 (2022-05-09) HAT — 38.73 (2022-05-09) HAT-L — 38.91 (2022-05-09) HAT — 38.73 (2022-05-09) HyperRes — 36.69 (2022-06-13) HyperRes — 36.69 (2022-06-13) HyperRes — 36.69 (2022-06-13) HyperRes — 36.69 (2022-06-13) HAT_FIR — 38.74 (2022-08-24) SwinFIR — 38.65 (2022-08-24) HAT_FIR — 38.74 (2022-08-24) SwinFIR — 38.65 (2022-08-24) HAT_FIR — 38.74 (2022-08-24) SwinFIR — 38.65 (2022-08-24) HAT_FIR — 38.74 (2022-08-24) SwinFIR — 38.65 (2022-08-24) FACD — 38.242 (2022-11-29) FACD — 38.242 (2022-11-29) FACD — 38.242 (2022-11-29) FACD — 38.242 (2022-11-29) SwinOIR — 38.21 (2023-03-16) SwinOIR — 38.21 (2023-03-16) SwinOIR — 38.21 (2023-03-16) SwinOIR — 38.21 (2023-03-16) DRCT-L — 39.14 (2024-03-31) DRCT — 38.72 (2024-03-31) DRCT-L — 39.14 (2024-03-31) DRCT — 38.72 (2024-03-31) DRCT-L — 39.14 (2024-03-31) DRCT — 38.72 (2024-03-31) DRCT-L — 39.14 (2024-03-31) DRCT — 38.72 (2024-03-31) HMA† — 38.95 (2024-05-08) HMA† — 38.95 (2024-05-08) HMA† — 38.95 (2024-05-08) HMA† — 38.95 (2024-05-08) CPAT+ — 38.72 (2024-07-23) CPAT — 38.68 (2024-07-23) CPAT+ — 38.72 (2024-07-23) CPAT — 38.68 (2024-07-23) CPAT+ — 38.72 (2024-07-23) CPAT — 38.68 (2024-07-23) CPAT+ — 38.72 (2024-07-23) CPAT — 38.68 (2024-07-23) ML-CrAIST — 38.19 (2024-08-19) ML-CrAIST-Li — 38.15 (2024-08-19) ML-CrAIST — 38.19 (2024-08-19) ML-CrAIST-Li — 38.15 (2024-08-19) ML-CrAIST — 38.19 (2024-08-19) ML-CrAIST-Li — 38.15 (2024-08-19) ML-CrAIST — 38.19 (2024-08-19) ML-CrAIST-Li — 38.15 (2024-08-19) Hi-IR-L — 38.87 (2024-11-27) Hi-IR-L — 38.87 (2024-11-27) Hi-IR-L — 38.87 (2024-11-27) Hi-IR-L — 38.87 (2024-11-27) MaIR+ — 38.62 (2024-12-28) MaIR — 38.56 (2024-12-28) MaIR+ — 38.62 (2024-12-28) MaIR — 38.56 (2024-12-28) MaIR+ — 38.62 (2024-12-28) MaIR — 38.56 (2024-12-28) MaIR+ — 38.62 (2024-12-28) MaIR — 38.56 (2024-12-28) DRCN [[Kim et al.2016b]] — 37.63 (2015-11-14) RED30 — 37.66 (2016-06-29) CARN [[Ahn et al.2018]] — 37.76 (2018-03-23) MWCNN — 37.91 (2018-05-18) SRFBN — 38.11 (2019-03-23) HBPN — 38.13 (2019-06-17) DRLN+ — 38.34 (2019-06-28) EDT-B — 38.63 (2021-12-19) HAT-L — 38.91 (2022-05-09) DRCT-L — 39.14 (2024-03-31)
RankModel PSNRSSIM Extra Training Data PaperCodeYear
1 DRCT-L 39.140.9658 DRCT: Saving Image Super-resolution away from Information Bottleneck ming053l/drct 2024
2 HMA† 38.950.9649 ✓ HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution korouuuuu/hma 2024
3 HAT-L 38.910.9646 Activating More Pixels in Image Super-Resolution Transformer xpixelgroup/hat · chxy95/hat 2022
4 Hi-IR-L 38.870.9663 Hierarchical Information Flow for Generalized Efficient Image Restoration 2024
5 HAT_FIR 38.74– SwinFIR: Revisiting the SwinIR with Fast Fourier Convolution and Improved Training for Image Super-Resolution Zdafeng/SwinFIR · IMPLabUniPr/swin2-mose 2022
6 HAT 38.730.9637 Activating More Pixels in Image Super-Resolution Transformer xpixelgroup/hat · chxy95/hat 2022
7 DRCT 38.720.9646 DRCT: Saving Image Super-resolution away from Information Bottleneck ming053l/drct 2024
7 CPAT+ 38.720.9635 Channel-Partitioned Windowed Attention And Frequency Learning for Single Image Super-Resolution 2024
9 CPAT 38.680.9633 Channel-Partitioned Windowed Attention And Frequency Learning for Single Image Super-Resolution 2024
10 SwinFIR 38.650.9633 SwinFIR: Revisiting the SwinIR with Fast Fourier Convolution and Improved Training for Image Super-Resolution Zdafeng/SwinFIR · IMPLabUniPr/swin2-mose 2022
11 EDT-B 38.630.9632 ✓ On Efficient Transformer-Based Image Pre-training for Low-Level Vision fenglinglwb/edt 2021
12 MaIR+ 38.620.963 MaIR: A Locality- and Continuity-Preserving Mamba for Image Restoration XLearning-SCU/2025-CVPR-MaIR 2024
13 MaIR 38.560.9628 MaIR: A Locality- and Continuity-Preserving Mamba for Image Restoration XLearning-SCU/2025-CVPR-MaIR 2024
14 DRLN+ 38.340.9619 Densely Residual Laplacian Super-Resolution saeed-anwar/DRLN 2019
15 HAN+ 38.330.9299 Single Image Super-Resolution via a Holistic Attention Network wwlCape/HAN · 04RR/SOTA-Vision 2020
15 LTE 38.33– Local Texture Estimator for Implicit Representation Function jaewon-lee-b/lte 2021
17 CSNLN 38.280.9616 Image Super-Resolution with Cross-Scale Non-Local Attention and Exhaustive Self-Exemplars Mining SHI-Labs/Cross-Scale-Non-Local-Attention · Lornatang/CSNLN-PyTorch · nSamsow/CSNLN-new 2020
18 FACD 38.242– Feature-domain Adaptive Contrastive Distillation for Efficient Single Image Super-Resolution 2022
19 PMRN+ 38.220.9612 Sequential Hierarchical Learning with Distribution Transformation for Image Super-Resolution 2020
20 SwinOIR 38.210.9614 Resolution Enhancement Processing on Low Quality Images Using Swin Transformer Based on Interval Dense Connection Strategy rubbbbbbbbby/swinoir · Mind23-2/MindCode-101 2023
1–20 / 164 다음 → 페이지당 10 20 50 100