List your company

Super-Resolution on Manga109

41.13PSNRHMA
35.09836.66438.2339.796Apr 12, 2017Oct 18, 2018Apr 24, 2020Oct 31, 2021May 8, 2023Nov 12, 2024May 21, 2026

Evaluation results

MethodLinks
HMAScale=x2, Pre-training strategy=true2024.0541.130.9836———
HAT-LScale=x2, Pre-training strategy=true2024.0541.010.9831———
SST-L+scale=×2, #params=20.3, Training dataset=DFLIP, training patch size=96 × 962026.0340.860.9821———
MambaIRV2-Lscale=×2, #params=34.2, Training dataset=DFLIP2026.0340.820.982———
HMAScale=x2, Pre-training strategy=false2024.0540.730.9824———
GRLScale=x2, Pre-training strategy=false2024.0540.670.9818———
PFTscale=×2, #params=19.6, Training dataset=DFLIP2026.0340.640.9817———
SST-Lscale=×2, #params=20.3, Training dataset=DFLIP2026.0340.640.9817———
HATScale=x2, Pre-training strategy=false2024.0540.260.9809———
SRFormerScale=x2, Pre-training strategy=false2024.0540.070.9802———
EDTScale=x2, Pre-training strategy=false2024.0539.930.98———
SwinIRScale=x2, Pre-training strategy=false2024.0539.920.9797———
IET-lightScale=x2, Params=783K, FLOPs=277.9G2026.0139.750.9793———
NLSAScale=x2, Pre-training strategy=false2024.0539.590.9789———
PFT-lightScale=x22025.0339.550.9792———
PFT-lightScale=x2, Params=776K, FLOPs=278.3G2026.0139.550.9792———
ATD-lightScale=x2, Params=753K2024.0139.510.9789———
ATD-lightScale=x22025.0339.510.9789———
ATD-lightScale=x2, Params=753K, FLOPs=348.6G2026.0139.510.9789———
RCANScale=x2, Pre-training strategy=false2024.0539.440.9786———
IGNNScale=x2, Pre-training strategy=false2024.0539.350.9786———
MambaIRv2-lightScale=x2, Params=774K, FLOPs=286.3G2026.0139.350.9785———
SANScale=x2, Pre-training strategy=false2024.0539.320.9792———
IPG-TinyScale=x22025.0339.310.9786———
IPG-TinyScale=x2, Params=872K, FLOPs=245.2G2026.0139.310.9786———
HBPNScale=2x2019.0639.30.979———
OmniSRScale=x2, Params=772K2024.0139.280.9784———
OmniSRScale=x22025.0339.280.9784———
OmniSRScale=x2, Params=772K, FLOPs=194.5G2026.0139.280.9784———
SwinIR-NGScale=x2, Params=1181K2024.0139.20.9781———
SwinIR-NGScale=x22025.0339.20.9781———
SwinIR-NGScale=x2, Params=1,181K, FLOPs=274.1G2026.0139.20.9781———
Omni-SRScale=x2, Params=772K, FLOPS=172.1G2024.0139.140.9777———
SwinIR-lightScale=x2, Params=910K2024.0139.120.9783———
DLGSANetScale=x2, Params=566K, FLOPS=128.1G2024.0139.120.9783———
SwinIR-lightScale=x22025.0339.120.9783———
SwinIR-lightScale=x2, Params=910K, FLOPs=244G2026.0139.120.9783———
ELANScale=x2, Params=582K2024.0139.110.9782———
ELANScale=x22025.0339.110.9782———
ELANScale=x2, Params=582K, FLOPs=203G2026.0139.110.9782———
EDSRScale=2x2019.0639.10.977———
DBPNScale=2x2019.0639.10.978———
EDSRScale=x2, Pre-training strategy=false2024.0539.10.9773———
SwinIR-lightScale=x2, Params=878K, FLOPS=195.6G2024.01390.9778———
CFSRScale=x2, Params=291K, FLOPS=62.6G2024.0138.940.9777———
IMDNScale=x2, Params=694K2024.0138.880.9774———
IMDNScale=x2, Params=694K, FLOPS=158.8G2024.0138.880.9774———
RFDNScale=x2, Params=534K, FLOPS=95.0G2024.0138.880.9773———
IMDNScale=x22025.0338.880.9774———
IMDNScale=x2, Params=694K, FLOPs=158.8G2026.0138.880.9774———
SPANScale=x2, Params=481K, FLOPS=94.4G2024.0138.830.9774———
PANScale=x2, Params=261K, FLOPS=70.5G2024.0138.70.9773———
LAPAR-AScale=x2, Params=548K2024.0138.670.9772———
LAPAR-AScale=x2, Params=548K, FLOPS=171.0G2024.0138.670.9772———
LAPAR-AScale=x22025.0338.670.9772———
LAPAR-AScale=x2, Params=548K, FLOPs=171G2026.0138.670.9772———
ShuffleMixerScale=x2, Params=394K, FLOPS=91.0G2024.0138.620.977———
SRResNetScale=x2, #Bit (w/a)=32/322026.0538.560.977———
QuantSR+*Scale=x2, #Bit (w/a)=4/42026.0538.550.977———
I2NQ*Scale=x2, #Bit (w/a)=4/42026.0538.480.9768———
QuantSR+Scale=x2, #Bit (w/a)=4/42026.0538.460.9767———
CARNScale=x2, Params=1,592K2024.0138.360.9765———
CARNScale=x22025.0338.360.9765———
CARNScale=x2, Params=1,592K, FLOPs=222.8G2026.0138.360.9765———
DCLSScale=x22022.0238.310.974———
QuantSRScale=x2, #Bit (w/a)=4/42026.0538.250.9762———
DANv2Scale=x22022.0238.070.9734———
CADyQScale=x2, #Bit (w/a)=8/82026.0538.060.976———
QuantSR+Scale=x2, #Bit (w/a)=2/22026.0537.960.9756———
DRRNScale=2x2019.0637.920.976———
PAMSScale=x2, #Bit (w/a)=4/42026.0537.620.974———
QuantSRScale=x2, #Bit (w/a)=2/22026.0537.60.9745———
DRCNScale=2x2017.0437.570.9739.541——
CADyQScale=x2, #Bit (w/a)=4/42026.0537.310.974———
LapSRN (ours 2x)Scale=2x2017.0437.270.9749.481——
LapSRNScale=2x2019.0637.270.974———
DANv1Scale=x22022.0237.230.971———
VDSRScale=2x2017.0437.160.9749.12——
VDSRScale=2x2019.0637.160.974———
DoReFaScale=x2, #Bit (w/a)=8/82026.0537.010.945———
DoReFaScale=x2, #Bit (w/a)=4/42026.0536.950.944———
IKCScale=x22022.0236.930.9667———
CRFSRScale=2x2019.0636.780.97———
LapSRN (ours 8x)Scale=2x2017.0436.730.9728.933——
FSRCNNScale=2x2017.0436.620.9718.587——
DoReFaScale=x2, #Bit (w/a)=2/22026.0536.130.941———
SST-L+scale=×3, #params=20.7, Training dataset=DFLIP, training patch size=96 × 962026.0336.120.958———
HMAScale=x3, Pre-training strategy=true2024.0536.10.958———
HAT-LScale=x3, Pre-training strategy=true2024.0536.020.9576———
MambaIRV2-Lscale=×3, #params=34.2, Training dataset=DFLIP2026.0335.980.9574———
SST-Lscale=×3, #params=20.7, Training dataset=DFLIP2026.0335.90.957———
HMAScale=x3, Pre-training strategy=false2024.0535.820.9567———
PFTscale=×3, #params=19.8, Training dataset=DFLIP2026.0335.810.9567———
SelfExSRScale=2x2017.0435.780.9688.721——
SRCNNScale=2x2017.0435.720.9688.471——
SRCNNScale=2x2019.0635.720.968———
HATScale=x3, Pre-training strategy=false2024.0535.530.9552———
SCNScale=2x2017.0435.470.9667.601——
A+Scale=2x2017.0435.330.9678.906——
A+Scale=2x2019.0635.330.967———