Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/HMA†

HMA†

Reported on 120 benchmarks across 4 tasks · 1 paper · 48 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision60 results

Image Super-ResolutiononSet14 - 3x upscaling
SSIM· uses extra data· 2024-05-08
0.8585
best: 0.8616 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononSet5 - 4x upscaling
PSNR· uses extra data· 2024-05-08
33.38
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononSet5 - 4x upscaling
SSIM· uses extra data· 2024-05-08
0.9089
best: 0.9103 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononSet5 - 3x upscaling
PSNR· uses extra data· 2024-05-08
35.35
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononManga109 - 4x upscaling
PSNR· uses extra data· 2024-05-08
33.19
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononUrban100 - 2x upscaling
PSNR· uses extra data· 2024-05-08
35.24
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononManga109 - 3x upscaling
PSNR· uses extra data· 2024-05-08
36.1
best: 36.12 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononManga109 - 3x upscaling
SSIM· uses extra data· 2024-05-08
0.958
best: 0.9588 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononUrban100 - 3x upscaling
PSNR· uses extra data· 2024-05-08
31
best: 31.07 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononUrban100 - 3x upscaling
SSIM· uses extra data· 2024-05-08
0.8984
best: 0.902 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononBSD100 - 3x upscaling
PSNR· uses extra data· 2024-05-08
29.66
best: 29.67 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononBSD100 - 3x upscaling
SSIM· uses extra data· 2024-05-08
0.8196
best: 0.8256 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononSet14 - 3x upscaling
SSIM· uses extra data· 2024-05-08
0.8585
best: 0.8616 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononSet5 - 4x upscaling
PSNR· uses extra data· 2024-05-08
33.38
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononSet5 - 4x upscaling
SSIM· uses extra data· 2024-05-08
0.9089
best: 0.9103 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononSet5 - 3x upscaling
PSNR· uses extra data· 2024-05-08
35.35
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononManga109 - 4x upscaling
PSNR· uses extra data· 2024-05-08
33.19
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononUrban100 - 2x upscaling
PSNR· uses extra data· 2024-05-08
35.24
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononManga109 - 3x upscaling
PSNR· uses extra data· 2024-05-08
36.1
best: 36.12 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononManga109 - 3x upscaling
SSIM· uses extra data· 2024-05-08
0.958
best: 0.9588 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononUrban100 - 3x upscaling
PSNR· uses extra data· 2024-05-08
31
best: 31.07 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononUrban100 - 3x upscaling
SSIM· uses extra data· 2024-05-08
0.8984
best: 0.902 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononBSD100 - 3x upscaling
PSNR· uses extra data· 2024-05-08
29.66
best: 29.67 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononBSD100 - 3x upscaling
SSIM· uses extra data· 2024-05-08
0.8196
best: 0.8256 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononBSD100 - 2x upscaling
PSNR· uses extra data· 2024-05-08
32.79
best: 33.12 (WaveMixSR-V2)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononBSD100 - 2x upscaling
SSIM· uses extra data· 2024-05-08
0.9071
best: 0.9326 (WaveMixSR-V2)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononSet14 - 3x upscaling
PSNR· uses extra data· 2024-05-08
31.47
best: 31.55 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononSet14 - 2x upscaling
PSNR· uses extra data· 2024-05-08
35.33
best: 35.36 (DRCT-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononSet14 - 2x upscaling
SSIM· uses extra data· 2024-05-08
0.9297
best: 0.9311 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononSet14 - 4x upscaling
PSNR· uses extra data· 2024-05-08
29.51
best: 29.54 (DRCT-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononSet14 - 4x upscaling
SSIM· uses extra data· 2024-05-08
0.8019
best: 0.894 (Edge-informed SR)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononSet5 - 3x upscaling
SSIM· uses extra data· 2024-05-08
0.9347
best: 0.938 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononManga109 - 4x upscaling
SSIM· uses extra data· 2024-05-08
0.9344
best: 0.9366 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononUrban100 - 2x upscaling
SSIM· uses extra data· 2024-05-08
0.9513
best: 0.9516 (DRCT-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononSet5 - 2x upscaling
PSNR· uses extra data· 2024-05-08
38.95
best: 39.14 (DRCT-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononSet5 - 2x upscaling
SSIM· uses extra data· 2024-05-08
0.9649
best: 0.9663 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononManga109 - 2x upscaling
PSNR· uses extra data· 2024-05-08
41.13
best: 41.22 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononManga109 - 2x upscaling
SSIM· uses extra data· 2024-05-08
0.9836
best: 0.9846 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononUrban100 - 4x upscaling
PSNR· uses extra data· 2024-05-08
28.69
best: 28.72 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononUrban100 - 4x upscaling
SSIM· uses extra data· 2024-05-08
0.8512
best: 0.9481 (SPSR)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononBSD100 - 4x upscaling
PSNR· uses extra data· 2024-05-08
28.13
best: 28.16 (DRCT-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Image Super-ResolutiononBSD100 - 4x upscaling
SSIM· uses extra data· 2024-05-08
0.7562
best: 0.851 (Edge-informed SR)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononBSD100 - 2x upscaling
PSNR· uses extra data· 2024-05-08
32.79
best: 33.12 (WaveMixSR-V2)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononBSD100 - 2x upscaling
SSIM· uses extra data· 2024-05-08
0.9071
best: 0.9326 (WaveMixSR-V2)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononSet14 - 3x upscaling
PSNR· uses extra data· 2024-05-08
31.47
best: 31.55 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononSet14 - 2x upscaling
PSNR· uses extra data· 2024-05-08
35.33
best: 35.36 (DRCT-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononSet14 - 2x upscaling
SSIM· uses extra data· 2024-05-08
0.9297
best: 0.9311 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononSet14 - 4x upscaling
PSNR· uses extra data· 2024-05-08
29.51
best: 29.54 (DRCT-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononSet14 - 4x upscaling
SSIM· uses extra data· 2024-05-08
0.8019
best: 0.894 (Edge-informed SR)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononSet5 - 3x upscaling
SSIM· uses extra data· 2024-05-08
0.9347
best: 0.938 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononManga109 - 4x upscaling
SSIM· uses extra data· 2024-05-08
0.9344
best: 0.9366 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononUrban100 - 2x upscaling
SSIM· uses extra data· 2024-05-08
0.9513
best: 0.9516 (DRCT-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononSet5 - 2x upscaling
PSNR· uses extra data· 2024-05-08
38.95
best: 39.14 (DRCT-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononSet5 - 2x upscaling
SSIM· uses extra data· 2024-05-08
0.9649
best: 0.9663 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononManga109 - 2x upscaling
PSNR· uses extra data· 2024-05-08
41.13
best: 41.22 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononManga109 - 2x upscaling
SSIM· uses extra data· 2024-05-08
0.9836
best: 0.9846 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononUrban100 - 4x upscaling
PSNR· uses extra data· 2024-05-08
28.69
best: 28.72 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononUrban100 - 4x upscaling
SSIM· uses extra data· 2024-05-08
0.8512
best: 0.9481 (SPSR)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononBSD100 - 4x upscaling
PSNR· uses extra data· 2024-05-08
28.13
best: 28.16 (DRCT-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
3D Object Super-ResolutiononBSD100 - 4x upscaling
SSIM· uses extra data· 2024-05-08
0.7562
best: 0.851 (Edge-informed SR)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001

Graphs30 results

Super-ResolutiononSet14 - 3x upscaling
SSIM· uses extra data· 2024-05-08
0.8585
best: 0.8616 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononSet5 - 4x upscaling
PSNR· uses extra data· 2024-05-08
33.38
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononSet5 - 4x upscaling
SSIM· uses extra data· 2024-05-08
0.9089
best: 0.9103 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononSet5 - 3x upscaling
PSNR· uses extra data· 2024-05-08
35.35
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononManga109 - 4x upscaling
PSNR· uses extra data· 2024-05-08
33.19
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononUrban100 - 2x upscaling
PSNR· uses extra data· 2024-05-08
35.24
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononManga109 - 3x upscaling
PSNR· uses extra data· 2024-05-08
36.1
best: 36.12 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononManga109 - 3x upscaling
SSIM· uses extra data· 2024-05-08
0.958
best: 0.9588 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononUrban100 - 3x upscaling
PSNR· uses extra data· 2024-05-08
31
best: 31.07 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononUrban100 - 3x upscaling
SSIM· uses extra data· 2024-05-08
0.8984
best: 0.902 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononBSD100 - 3x upscaling
PSNR· uses extra data· 2024-05-08
29.66
best: 29.67 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononBSD100 - 3x upscaling
SSIM· uses extra data· 2024-05-08
0.8196
best: 0.8256 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononBSD100 - 2x upscaling
PSNR· uses extra data· 2024-05-08
32.79
best: 33.12 (WaveMixSR-V2)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononBSD100 - 2x upscaling
SSIM· uses extra data· 2024-05-08
0.9071
best: 0.9326 (WaveMixSR-V2)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononSet14 - 3x upscaling
PSNR· uses extra data· 2024-05-08
31.47
best: 31.55 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononSet14 - 2x upscaling
PSNR· uses extra data· 2024-05-08
35.33
best: 35.36 (DRCT-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononSet14 - 2x upscaling
SSIM· uses extra data· 2024-05-08
0.9297
best: 0.9311 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononSet14 - 4x upscaling
PSNR· uses extra data· 2024-05-08
29.51
best: 29.54 (DRCT-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononSet14 - 4x upscaling
SSIM· uses extra data· 2024-05-08
0.8019
best: 0.894 (Edge-informed SR)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononSet5 - 3x upscaling
SSIM· uses extra data· 2024-05-08
0.9347
best: 0.938 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononManga109 - 4x upscaling
SSIM· uses extra data· 2024-05-08
0.9344
best: 0.9366 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononUrban100 - 2x upscaling
SSIM· uses extra data· 2024-05-08
0.9513
best: 0.9516 (DRCT-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononSet5 - 2x upscaling
PSNR· uses extra data· 2024-05-08
38.95
best: 39.14 (DRCT-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononSet5 - 2x upscaling
SSIM· uses extra data· 2024-05-08
0.9649
best: 0.9663 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononManga109 - 2x upscaling
PSNR· uses extra data· 2024-05-08
41.13
best: 41.22 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononManga109 - 2x upscaling
SSIM· uses extra data· 2024-05-08
0.9836
best: 0.9846 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononUrban100 - 4x upscaling
PSNR· uses extra data· 2024-05-08
28.69
best: 28.72 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononUrban100 - 4x upscaling
SSIM· uses extra data· 2024-05-08
0.8512
best: 0.9481 (SPSR)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononBSD100 - 4x upscaling
PSNR· uses extra data· 2024-05-08
28.13
best: 28.16 (DRCT-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
Super-ResolutiononBSD100 - 4x upscaling
SSIM· uses extra data· 2024-05-08
0.7562
best: 0.851 (Edge-informed SR)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001

Methodology30 results

16konSet14 - 3x upscaling
SSIM· uses extra data· 2024-05-08
0.8585
best: 0.8616 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konSet5 - 4x upscaling
PSNR· uses extra data· 2024-05-08
33.38
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konSet5 - 4x upscaling
SSIM· uses extra data· 2024-05-08
0.9089
best: 0.9103 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konSet5 - 3x upscaling
PSNR· uses extra data· 2024-05-08
35.35
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konManga109 - 4x upscaling
PSNR· uses extra data· 2024-05-08
33.19
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konUrban100 - 2x upscaling
PSNR· uses extra data· 2024-05-08
35.24
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konManga109 - 3x upscaling
PSNR· uses extra data· 2024-05-08
36.1
best: 36.12 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konManga109 - 3x upscaling
SSIM· uses extra data· 2024-05-08
0.958
best: 0.9588 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konUrban100 - 3x upscaling
PSNR· uses extra data· 2024-05-08
31
best: 31.07 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konUrban100 - 3x upscaling
SSIM· uses extra data· 2024-05-08
0.8984
best: 0.902 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konBSD100 - 3x upscaling
PSNR· uses extra data· 2024-05-08
29.66
best: 29.67 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konBSD100 - 3x upscaling
SSIM· uses extra data· 2024-05-08
0.8196
best: 0.8256 (Hi-IR-L)
SOTA
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konBSD100 - 2x upscaling
PSNR· uses extra data· 2024-05-08
32.79
best: 33.12 (WaveMixSR-V2)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konBSD100 - 2x upscaling
SSIM· uses extra data· 2024-05-08
0.9071
best: 0.9326 (WaveMixSR-V2)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konSet14 - 3x upscaling
PSNR· uses extra data· 2024-05-08
31.47
best: 31.55 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konSet14 - 2x upscaling
PSNR· uses extra data· 2024-05-08
35.33
best: 35.36 (DRCT-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konSet14 - 2x upscaling
SSIM· uses extra data· 2024-05-08
0.9297
best: 0.9311 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konSet14 - 4x upscaling
PSNR· uses extra data· 2024-05-08
29.51
best: 29.54 (DRCT-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konSet14 - 4x upscaling
SSIM· uses extra data· 2024-05-08
0.8019
best: 0.894 (Edge-informed SR)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konSet5 - 3x upscaling
SSIM· uses extra data· 2024-05-08
0.9347
best: 0.938 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konManga109 - 4x upscaling
SSIM· uses extra data· 2024-05-08
0.9344
best: 0.9366 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konUrban100 - 2x upscaling
SSIM· uses extra data· 2024-05-08
0.9513
best: 0.9516 (DRCT-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konSet5 - 2x upscaling
PSNR· uses extra data· 2024-05-08
38.95
best: 39.14 (DRCT-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konSet5 - 2x upscaling
SSIM· uses extra data· 2024-05-08
0.9649
best: 0.9663 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konManga109 - 2x upscaling
PSNR· uses extra data· 2024-05-08
41.13
best: 41.22 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konManga109 - 2x upscaling
SSIM· uses extra data· 2024-05-08
0.9836
best: 0.9846 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konUrban100 - 4x upscaling
PSNR· uses extra data· 2024-05-08
28.69
best: 28.72 (Hi-IR-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konUrban100 - 4x upscaling
SSIM· uses extra data· 2024-05-08
0.8512
best: 0.9481 (SPSR)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konBSD100 - 4x upscaling
PSNR· uses extra data· 2024-05-08
28.13
best: 28.16 (DRCT-L)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001
16konBSD100 - 4x upscaling
SSIM· uses extra data· 2024-05-08
0.7562
best: 0.851 (Edge-informed SR)
HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution arXiv:2405.05001