List your company

Multi-person Pose Estimation on COCO 2017 (test-dev)

78.1APMSPN
33.79645.29856.868.302Nov 16, 2016Apr 22, 2018Sep 26, 2019Mar 2, 2021Aug 6, 2022Jan 10, 2024Jun 16, 2025

Evaluation results

MethodLinks
MSPNBackbone=4xRes-50, Input Size=384×288, Ensemble=true, External Data=true2019.0178.194.185.974.583.383.196.789.879.388.2
MSPNBackbone=4xRes-50, Input Size=384×288, External Data=true, Ensemble=false2019.0177.193.884.673.482.382.396.588.978.487.7
HRNetBackbone=HRNet-W48, Input Size=384×288, External Data=true2019.017792.784.573.483.182————
Simple BaseBackbone=Res-152, Input Size=384×288, Ensemble=true, External Data=true2019.0176.592.4847382.781.595.888.277.487.2
MSPNBackbone=4xRes-50, Input Size=384×288, Ensemble=false, External Data=false2019.0176.193.483.872.381.581.696.388.177.587.1
HRNetBackbone=HRNet-W48, Input Size=384×2882019.0175.592.583.371.981.580.5————
HRNet-W48category=Top-down2020.0775.592.583.371.981.580.5————
GroupPoseBackbone=Swin-L, #Params=219.6M, Time (ms)=214, End-to-end=true2025.0674.891.682.169.483—————
Simple BaseBackbone=Res-152, Input Size=384×2882019.0173.791.981.170.38079————
SBLcategory=Top-down2020.0773.791.981.170.38079————
QueryPoseBackbone=Swin-L, End-to-end=true2025.0673.391.379.568.581.2—————
RTMO-lBackbone=CSPDarknet, #Params=44.8M, Time (ms)=19.1, additional_training_data=true2023.1273.391.9—68.381.177.4————
CPNBackbone=Res-Inception, Input Size=384×288, Ensemble=true2019.017391.780.969.578.17995.185.974.884.6
ED-PoseBackbone=Swin-L, #Params=218.0M, Time (ms)=222.6, End-to-end=true2025.0672.792.380.967.680—————
ED-PoseBackbone=Swin-L, #Params=218.0M, Time (ms)=265.62023.1272.792.380.967.680—————
CFNcategory=Top-down2020.0772.686.169.778.364.1—————
RMPEcategory=Top-down2020.0772.389.279.16878.6—————
DETRPose-XBackbone=HGNetv2-B5, #Params=82.3M, Time (ms)=52.1, End-to-end=true2025.0672.291.479.36780.678.8————
CPNBackbone=Res-Inception, Input Size=384×2882019.0172.191.48068.777.278.595.185.374.284.3
CPNcategory=Top-down2020.0772.191.48068.777.278.5————
RTMO-LBackbone=CSPDarknet, #Params=44.8M, Time (ms)=19.1, End-to-end=false2025.0671.691.17966.879.175.6————
RTMO-lBackbone=CSPDarknet, #Params=44.8M, Time (ms)=19.12023.1271.691.17966.879.175.6————
RTMO-mBackbone=CSPDarknet, #Params=22.6M, Time (ms)=12.4, additional_training_data=true2023.1271.59178.666.179.975.6————
DETRPose-LBackbone=HGNetv2-B4, #Params=36.8M, Time (ms)=32.5, End-to-end=true2025.0671.291.278.165.879.978.1————
RTMO-r50Backbone=ResNet-50, #Params=41.7M, Time (ms)=15.5, End-to-end=false2025.0670.99178.265.879.175————
RTMO-r50Backbone=ResNet-50, #Params=41.7M, Time (ms)=15.52023.1270.99178.265.879.175————
CIDBackbone=HRNet-w48, #Params=65.4M, Time (ms)=94.8, End-to-end=false2025.0670.790.477.966.377.876.4————
CIDBackbone=HRNet-w48, #Params=65.4M, Time (ms)=94.82023.1270.790.477.966.377.876.4————
PETRBackbone=Swin-L, #Params=213.8M, Time (ms)=133, End-to-end=true2025.0670.591.578.765.278—————
PETRBackbone=Swin-L, #Params=213.8M, Time (ms)=1332023.1270.591.578.765.278—————
KAPAO-LBackbone=CSPNet, #Params=77.0M, Time (ms)=50.2, End-to-end=false2025.0670.391.277.866.376.877.7————
KAPAO-lBackbone=CSPNet, #Params=77.0M, Time (ms)=50.22023.1270.391.277.866.376.877.7————
YOLO-Pose-lBackbone=CSPDarknet, #Params=61.3M, Time (ms)=20.5, End-to-end=false2025.0670.291.177.865.378.274.3————
GroupPoseBackbone=ResNet-50, #Params=52.4M, Time (ms)=116.5, End-to-end=true2025.0670.290.577.864.778—————
YOLO-Pose-lBackbone=CSPDarknet, #Params=61.3M, Time (ms)=20.52023.1270.291.177.865.378.274.3————
RTMO-MBackbone=CSPDarknet, #Params=22.6M, Time (ms)=12.4, End-to-end=false2025.0670.190.677.165.178.174.2————
RTMO-mBackbone=CSPDarknet, #Params=22.6M, Time (ms)=12.42023.1270.190.677.165.178.174.2————
ED-PoseBackbone=ResNet-50, #Params=50.6M, Time (ms)=125.5, End-to-end=true2025.0669.890.277.264.377.4—————
ED-PoseBackbone=ResNet-50, #Params=50.6M, Time (ms)=135.22023.1269.890.277.264.377.4—————
CIDBackbone=HRNet-w32, #Params=29.4M, Time (ms)=84, End-to-end=false2025.0668.989.976.963.277.774.6————
CIDBackbone=HRNet-w32, #Params=29.4M, Time (ms)=84.02023.1268.989.976.963.277.774.6————
KAPAO-MBackbone=CSPNet, #Params=35.8M, Time (ms)=37, End-to-end=false2025.0668.890.576.564.37676.3————
KAPAO-mBackbone=CSPNet, #Params=35.8M, Time (ms)=37.02023.1268.890.576.564.37676.3————
QueryPoseBackbone=ResNet-50, End-to-end=true2025.0668.788.674.463.876.5—————
PersonLabcategory=Bottom-up, multi-scale test=true2020.0768.78975.464.175.575.4————
YOLO-Pose-mBackbone=CSPDarknet, #Params=29.3M, Time (ms)=12.5, End-to-end=false2025.0668.690.775.863.477.172.8————
YOLO-Pose-mBackbone=CSPDarknet, #Params=29.3M, Time (ms)=12.52023.1268.690.775.863.477.172.8————
DETRPose-MBackbone=HGNetv2-B2, #Params=23.5M, Time (ms)=25.8, End-to-end=true2025.0668.588.674.662.178.374.6————
HigherHRNet-W48Input Size=640x640, #Params=63.8M, #MACs=155.1G, Nano Latency (ms)=2101, Mobile Latency (ms)=1532, Pi Latency (ms)=123022022.0568.4—————————
IPRcategory=Top-down2020.0767.888.274.863.974—————
RTMO-sBackbone=CSPDarknet, #Params=9.9M, Time (ms)=8.9, additional_training_data=true2023.1267.789.474.561.577.271.9————
PETRBackbone=ResNet-50, #Params=43.7M, Time (ms)=89, End-to-end=true2025.0667.689.875.361.676—————
HGGcategory=Bottom-up, multi-scale test=true, single-person pose refinement=false2020.0767.685.173.762.774.671.3————
PETRBackbone=ResNet-50, #Params=43.7M, Time (ms)=892023.1267.689.875.361.676—————
RTMO-SBackbone=CSPDarknet, #Params=9.9M, Time (ms)=8.9, End-to-end=false2025.0666.988.873.661.175.770.9————
SPMcategory=Bottom-up, single-person pose refinement=true, multi-scale test=true2020.0766.988.572.962.673.1—————
RTMO-sBackbone=CSPDarknet, #Params=9.9M, Time (ms)=8.92023.1266.988.873.661.175.770.9————
DETRPose-SBackbone=HGNetv2-B0, #Params=11.9M, Time (ms)=20.3, End-to-end=true2025.0666.587.472.559.876.873————
PersonLabInput Size=1401x1401, #Params=68.7M, #MACs=405.5G2022.0566.5—————————
InsPoseBackbone=ResNet-101, Time (ms)=100, End-to-end=false2025.0666.389.27361.273.9—————
InsPoseBackbone=ResNet-101, Time (ms)=1002023.1266.389.27361.273.9—————
FCPoseBackbone=ResNet-101, #Params=60.5M, Time (ms)=93, End-to-end=false2025.0665.687.972.662.172.3—————
FCPoseBackbone=ResNet-101, #Params=60.5M, Time (ms)=932023.1265.687.972.662.172.3—————
Associative Embeddingtraining data=excluding external data2016.1165.586.872.360.672.670.289.57664.678.1
AEBackbone=Not specified, Input Size=512×5122019.0165.586.872.360.672.670.289.57664.678.1
AEcategory=Bottom-up, single-person pose refinement=true, multi-scale test=true2020.0765.586.872.360.672.670.2————
InsPoseBackbone=ResNet-50, #Params=50.2M, Time (ms)=80, End-to-end=false2025.0665.488.971.760.272.7—————
InsPoseBackbone=ResNet-50, #Params=50.2M, Time (ms)=802023.1265.488.971.760.272.7—————
G-RMItraining data=excluding external data2016.1164.985.571.362.37069.788.775.564.477.1
G-RMIBackbone=Res-152, Input Size=353×2572019.0164.985.571.362.37069.788.775.564.477.1
G-RMIcategory=Top-down2020.0764.985.571.362.37069.7————
G-RMI PoseInitialization=JFT-300M2017.0764.885.8———69.488.4———
Directposecategory=Bottom-up, multi-scale test=true2020.0764.887.871.160.471.5—————
G-RMI PoseInitialization=ImageNet + JFT-300M2017.0764.485.7———69.188.2———
FCPoseBackbone=ResNet-50, #Params=41.7M, Time (ms)=68, End-to-end=false2025.0664.387.37161.670.5—————
FCPoseBackbone=ResNet-50, #Params=41.7M, Time (ms)=682023.1264.387.37161.670.5—————
KAPAO-SBackbone=CSPNet, #Params=12.6M, Time (ms)=26.9, End-to-end=false2025.0663.888.470.458.671.771.2————
KAPAO-sBackbone=CSPNet, #Params=12.6M, Time (ms)=26.92023.1263.888.470.458.671.771.2————
DirectPoseBackbone=ResNet-101, End-to-end=false2025.0663.386.769.457.871.2—————
DirectPoseBackbone=ResNet-1012023.1263.386.769.457.871.2—————
YOLO-Pose-sBackbone=CSPDarknet, #Params=10.8M, Time (ms)=7.9, End-to-end=false2025.0663.287.869.557.672.667.6————
YOLO-Pose-sBackbone=CSPDarknet, #Params=10.8M, Time (ms)=7.92023.1263.287.869.557.672.667.6————
Mask R-CNNBackbone=Res-50-FPN, Input Size=Not specified2019.0163.187.368.757.871.4—————
Mask-RCNNcategory=Top-down2020.0763.187.368.757.871.469.7————
CenterNetBackbone=Hourglass, #Params=194.9M, Time (ms)=160, End-to-end=false2025.066386.869.658.970.4—————
CenterNetBackbone=Hourglass, #Params=194.9M, Time (ms)=1602023.126386.869.658.970.4—————
Mask-RCNNtraining data=excluding external data2016.1162.78768.457.471.1—————
G-RMI PoseInitialization=ImageNet2017.0762.484———66.786.6———
DirectPoseBackbone=ResNet-50, Time (ms)=74, End-to-end=false2025.0662.286.468.256.769.8—————
DirectPoseBackbone=ResNet-50, Time (ms)=742023.1262.286.468.256.769.8—————
CMU-Posetraining data=excluding external data2016.1161.884.967.557.168.266.587.271.860.674.6
CMU Pose2017.0761.884.9———66.587.2———
CMU PoseBackbone=Not specified, Input Size=Not specified2019.0161.884.967.557.168.266.587.271.860.674.6
OpenPosecategory=Bottom-up, single-person pose refinement=true2020.0761.884.967.557.168.266.5————
LitePose-SInput Size=448x448, #Params=2.7M, #MACs=5.0G, Nano Latency (ms)=97, Mobile Latency (ms)=76, Pi Latency (ms)=4202022.0556.7—————————
HourglassInput Size=512x512, #Params=277.8M, #MACs=206.9G2022.0556.6—————————
EfficientHRNet-H-2Input Size=448x448, #Params=8.3M, #MACs=7.9G, Nano Latency (ms)=170, Mobile Latency (ms)=182, Pi Latency (ms)=8782022.0552.8—————————
LitePose-XSInput Size=256x256, #Params=1.7M, #MACs=1.2G, Nano Latency (ms)=22, Mobile Latency (ms)=27, Pi Latency (ms)=1092022.0537.8—————————
EfficientHRNet-H-4Input Size=384x384, #Params=2.8M, #MACs=2.2G, Nano Latency (ms)=50, Mobile Latency (ms)=78, Pi Latency (ms)=2732022.0535.5—————————