List your company

Human Pose Estimation on COCO (test-dev)

81.1APViTPose
74.96476.55778.1579.743Dec 18, 2018Jan 13, 2020Feb 7, 2021Mar 6, 2022Apr 1, 2023Apr 26, 2024May 23, 2025

Evaluation results

MethodLinks
ViTPoseBackbone=VITAE-G, Ensemble=true2022.0481.19588.277.88685.6——————————
ViTPoseBackbone=VITAE-G, Ensemble=false2022.0480.994.888.177.585.985.4——————————
UDP+++Backbone=HRNet-w48plus, Ensemble=true2022.0480.894.988.177.485.785.3——————————
SCAIBackbone=HR48, Size=384×2882023.0380.694.88778.184.883.1——————————
UDP-Pose-PSA(p)Backbone=HRNet-W48, Input Size=384 x 288, Flops=35.4G, mPara=70.1M2021.0779.593.685.976.384.381.9——————————
UDP-Pose-PSA(s)Backbone=HRNet-W48, Input Size=384 x 288, Flops=35.4G, mPara=69.1M2021.0779.493.685.876.184.181.7——————————
SCIOBackbone=HR48, Size=384×2882022.0779.293.585.874.184.281.6——————————
RSN++Backbone=4xRSN-50, Ensemble=false2022.0479.294.487.176.183.884.1——————————
SCIOBackbone=HR48, Size=384×2882023.0379.293.585.874.184.281.6——————————
UDP-Pose-PSA(p)Backbone=HRNet-W48, Input Size=256 x 192, Flops=15.7G, mPara=70.1M2021.0778.993.685.876.183.681.4——————————
CCMBackbone=HRNet-w48, Ensemble=true2022.0478.993.8867584.583.6——————————
UDPv1*+AIDBackbone=HRNet-W48plus, Input size=384 x 288, Paradigm=Top-down, Extra data=true2020.0878.793.185.975.284.584——————————
PoseBHBackbone=Huge2025.0578.693.386.275.384.483.5——————————
ViTPose++Backbone=Huge2025.0578.593.486.275.384.483.4——————————
PoseurBackbone=HRFormer-B, Paradigm=Regression-based, Number of Decoders=6, Input Resolution=384 x 2882022.0178.393.585.975.283.4———————————
UDPv1*Backbone=HRNet-W48plus, Input size=384 x 288, Paradigm=Top-down, Extra data=true2020.0878.292.885.474.884.283.6——————————
MegviiApproach Type=Top-Down2018.1278.194.185.974.583.3———————————
MSPN++Backbone=4xResNet-50, Ensemble=false2022.0478.194.185.974.583.383.1——————————
UDP-Pose (Strong Baseline)Backbone=HRNet-W48, Input Size=384 x 288, Flops=33.0G, mPara=63.8M, Extra training data=Yes2021.0777.89284.374.284.582.5——————————
PoseurBackbone=HRNet-W48, Paradigm=Regression-based, Number of Decoders=6, Input Resolution=384 x 2882022.0177.692.98574.482.7———————————
UDPv1+AIDBackbone=HRNet-W48plus, Input size=384 x 288, Paradigm=Top-down2020.0877.592.984.974.283.483——————————
Ours (+DARK)Network=HRNetW48, Input Size=384 x 2882023.0377.5————82.4——————————
DARKBackbone=HRNet-W48, Input size=384 × 288, #Params=63.6M, GFLOPs=32.9, Extra data=true2019.1077.492.684.673.683.782.3——————————
DARK*Backbone=HRNet-W48, Input size=384 x 288, Paradigm=Top-down, Extra data=true2020.0877.492.684.673.683.782.3——————————
DARKBackbone=HRNet-w48, Ensemble=false2022.0477.492.684.673.683.782.3——————————
UDP-PoseBackbone=HRNet-W48, Input Size=256 x 192, Flops=14.7G, mPara=63.8M2021.0777.291.883.773.883.782——————————
DUAL (+DARK)Network=HRNetW48, Input Size=384 x 2882023.0377.2————82.2——————————
MSPNBackbone=MSPN, Input size=384 x 288, Paradigm=Top-down, Extra data=true2020.0877.193.884.673.482.382.3——————————
HRNet-W48Backbone=HRNet-W48, Input size=384 x 288, #Params=63.6M, GFLOPS=32.9, Extra data=extra data, Paradigm=Top-down2019.027792.784.573.483.182——————————
HRNetBackbone=HRNet-W48, Input size=384 × 288, #Params=63.6M, GFLOPs=32.9, Extra data=true2019.107792.784.573.483.182——————————
HRNet*Backbone=HRNet-W48, Input size=384 x 288, Paradigm=Top-down, Extra data=true2020.087792.784.573.483.182——————————
UDPv1+AIDBackbone=HRNet-W32, Input size=384 x 288, Paradigm=Top-down2020.087792.984.573.682.882.5——————————
HRNetBackbone=HRNet-w48, Ensemble=false2022.047792.784.573.483.182——————————
Graph-PCNNBackbone=HRNet-W48, Input size=384 x 288, Paradigm=Top-down2020.0876.892.684.373.382.781.6——————————
UDPv1Backbone=HRNet-W48plus, Input size=384 x 288, Paradigm=Top-down2020.0876.892.884.173.482.682.4——————————
Dark-PoseBackbone=HRNet-W48, Input Size=384 x 288, Flops=32.9G, mPara=63.6M2021.0776.890.683.272.88481.7——————————
Graph-PCNNBackbone=HR48, Size=384×2882022.0776.892.684.373.382.781.6——————————
Graph-PCNNBackbone=HR48, Size=384×2882023.0376.892.684.373.382.781.6——————————
PoseFixBackbone=HRNet-W48+ResNet-152, Input size=384 x 288, Paradigm=Top-down2020.0876.792.684.173.182.681.5——————————
UDPv1+AIDBackbone=HRNet-W48, Input size=256 x 192, Paradigm=Top-down2020.0876.792.984.273.482.382.3——————————
PoseFixBackbone=HR48+R152, Size=384×2882022.0776.792.684.173.182.681.5——————————
PoseFixBackbone=HR48+R152, Size=384×2882023.0376.792.684.173.182.681.5——————————
UDPv1+AIDBackbone=2xRSN-50, Input size=256 x 192, Paradigm=Top-down2020.0876.692.884.373.682.182.5——————————
PoseBHBackbone=Base2025.0576.692.684.473.482.481.7——————————
MRSAApproach Type=Top-Down2018.1276.592.4847382.7———————————
UDPv1Backbone=HRNet-W32, Input size=384 x 288, Paradigm=Top-down2020.0876.592.683.872.982.582.1——————————
HRNet + UDPBackbone=HRNet-W48, Input size=384 x 288, Paradigm=Top-down2019.1176.592.7847382.481.6——————————
UDPBackbone=HR48, Size=384×2882022.0776.592.7847382.481.6——————————
UDP-PoseBackbone=HRNet-W48, Paradigm=Heatmap-based, Input Resolution=384 x 2882022.0176.592.7847382.4———————————
BaselineBackbone=ResNet-152, Ensemble=true2022.0476.592.4847382.781.5——————————
UDPNetwork=HRNetW48, Input Size=384 x 2882023.0376.5————81.6——————————
UDPBackbone=HR48, Size=384×2882023.0376.592.7847382.481.6——————————
ViTPose++Backbone=Base2025.0576.492.784.373.282.281.5——————————
HRNet-W48Backbone=HRNet, Input Size=384 x 288, Flops=32.9G, mPara=63.6M2021.0776.390.882.972.383.481.2——————————
DARKBackbone=HRNet-W48, Input size=384 × 288, #Params=63.6M, GFLOPs=32.92019.1076.292.583.672.582.481.1——————————
DARKBackbone=HRNet-W48, Input size=384 x 288, Paradigm=Top-down2020.0876.292.583.672.582.481.1——————————
UDPv1+AIDBackbone=HRNet-W32, Input size=256 x 192, Paradigm=Top-down2020.0876.292.883.872.981.881.8——————————
DARKbackbone=HRNetV1-W48, input size=384 x 288, #Params=63.6M, GFLOPS=32.92021.0476.292.583.672.582.481.1——————————
DARKPrediction Type=Heatmap based, Backbone=HRNet-W48, Input size=384 × 288, #Params=63.6M, GFLOPS=32.9, Multi-scale test=false2021.0476.292.583.672.582.481.1——————————
UDP-PoseBackbone=HRNet-W48, Input Size=384 x 288, Flops=33.0G, mPara=63.8M2021.0776.292.583.672.582.481.1——————————
HRFormer-Binput size=384 x 288, #param.=43.2M, FLOPS=26.8G, Pre-training=ImageNet2021.1076.292.783.872.582.381.2——————————
DARKBackbone=HR48, Size=384×2882022.0776.292.583.672.582.481.1——————————
HRFormerBackbone=HRFormer-B, Paradigm=Heatmap-based, Input Resolution=384 x 2882022.0176.292.783.872.582.3———————————
DARKNetwork=HRNetW48, Input Size=384 x 2882023.0376.2————81.1——————————
DARKBackbone=HR48, Size=384×2882023.0376.292.583.672.582.481.1——————————
UDPv1Backbone=HRNet-W48, Input size=256 x 192, Paradigm=Top-down2020.0876.192.483.772.781.981.8——————————
MSPNBackbone=MSPN, Input size=384 x 288, Paradigm=Top-down2019.1176.193.483.872.381.581.6——————————
HRNet + UDPBackbone=HRNet-W32, Input size=384 x 288, Paradigm=Top-down2019.1176.192.583.572.88281.3——————————
MSPNBackbone=MSPN, Size=384×2882022.0776.193.483.872.381.581.6——————————
MSPNNetwork=ResNet50, Input Size=384 x 2882023.0376.1————81.6——————————
MSPNBackbone=MSPN, Size=384×2882023.0376.193.483.872.381.581.6——————————
UDPv1Backbone=2xRSN-50, Input size=256 x 192, Paradigm=Top-down2020.087692.583.673.281.482.3——————————
The Sea MonstersApproach Type=Top-Down2018.1275.992.18371.782.1———————————
TokenPose-L/D24input size=384 x 288, #param.=29.8M, FLOPS=22.1G, Pre-training=ImageNet2021.1075.992.383.472.282.180.8——————————
TokenPoseBackbone=L/D24, Paradigm=Heatmap-based, Input Resolution=384 x 2882022.0175.992.383.472.282.1———————————
HRNet-W32Backbone=HRNet, Input Size=384 x 288, Flops=16.0G, mPara=28.5M2021.0775.890.682.771.982.881——————————
HRNet + UDPBackbone=HRNet-W48, Input size=256 x 192, Paradigm=Top-down2019.1175.792.483.372.581.480.9——————————
EvoPose2D-LBackbone=ResNet-1012021.0775.791.983.172.281.5———————————
HRNet + RLEBackbone=HRNet-W482021.0775.792.382.972.381.3———————————
EvoPose2D-LStage=Two-stage, Test-Time Augmentation (TTA)=true2021.1175.791.983.172.281.581.7——————————
MIPNetStage=Two-stage, Test-Time Augmentation (TTA)=true2021.1175.7———————————————
RLEStage=Two-stage, Test-Time Augmentation (TTA)=true2021.1175.792.382.972.381.3———————————
RLEBackbone=HRNet-W48, Paradigm=Regression-based, Input Resolution=384 x 2882022.0175.792.382.972.381.3———————————
UDPv1Backbone=HRNet-W32, Input size=256 x 192, Paradigm=Top-down2020.0875.692.383.272.381.481.4——————————
Dark-PoseBackbone=HRNet-W32, Input Size=256 x 192, Flops=7.1G, mPara=28.5M2021.0775.690.582.171.882.880.8——————————
HRNet-W48Backbone=HRNet-W48, Input size=384 x 288, #Params=63.6M, GFLOPS=32.9, Paradigm=Top-down2019.0275.592.583.371.981.580.5——————————
HRNetBackbone=HRNet-W48, Input size=384 × 288, #Params=63.6M, GFLOPs=32.92019.1075.592.583.371.981.580.5——————————
HRNetBackbone=HRNet-W48, Input size=384 x 288, Paradigm=Top-down2020.0875.592.583.371.981.580.5——————————
RSNBackbone=2xRSN-50, Input size=256 x 192, Paradigm=Top-down2020.0875.593.6847379.681.3——————————
HRNetBackbone=HRNet-W48, Input size=384 x 288, Paradigm=Top-down2019.1175.592.583.371.981.580.5——————————
HRNetV1backbone=HRNetV1-W48, input size=384 x 288, #Params=63.6M, GFLOPS=32.92021.0475.592.583.371.981.580.5——————————
HRNetPrediction Type=Heatmap based, Backbone=HRNet-W48, Input size=384 × 288, #Params=63.6M, GFLOPS=32.9, Multi-scale test=false2021.0475.592.583.371.981.580.5——————————
HRNetBackbone=HRNet-W482021.0775.592.583.371.981.5———————————
HRNet-W48input size=384 x 288, #param.=63.6M, FLOPS=32.9G, Pre-training=ImageNet2021.1075.592.583.371.981.580.5——————————
HRNetBackbone=HR48, Size=384×2882022.0775.592.583.371.981.580.5——————————
HRNet-W48Stage=Two-stage, Test-Time Augmentation (TTA)=true2021.1175.592.583.371.981.580.5——————————
HRNetBackbone=HRNet-W48, Paradigm=Heatmap-based, Input Resolution=384 x 2882022.0175.592.583.371.981.5———————————
HRNetNetwork=HRNetW48, Input Size=384 x 2882023.0375.5————80.5——————————
HRNetBackbone=HR48, Size=384×2882023.0375.592.583.371.981.580.5——————————
HRNet + UDPBackbone=HRNet-W32, Input size=256 x 192, Paradigm=Top-down2019.1175.292.482.97280.880.4——————————