SOTAVerified

2D Human Pose Estimation

What is Human Pose Estimation? Human pose estimation is the process of estimating the configuration of the body (pose) from a single, typically monocular, image. Background. Human pose estimation is one of the key problems in computer vision that has been studied for well over 15 years. The reason for its importance is the abundance of applications that can benefit from such a technology. For example, human pose estimation allows for higher-level reasoning in the context of human-computer interaction and activity recognition; it is also one of the basic building blocks for marker-less motion capture (MoCap) technology. MoCap technology is useful for applications ranging from character animation to clinical analysis of gait pathologies.

Papers

Showing 125 of 118 papers

TitleStatusHype
Sapiens: Foundation for Human Vision ModelsCode9
Effective Whole-body Pose Estimation with Two-stages DistillationCode4
BlazePose: On-device Real-time Body Pose trackingCode4
XMem: Long-Term Video Object Segmentation with an Atkinson-Shiffrin Memory ModelCode3
Realtime Multi-Person 2D Pose Estimation using Part Affinity FieldsCode3
ViTPose: Simple Vision Transformer Baselines for Human Pose EstimationCode3
RMPE: Regional Multi-person Pose EstimationCode3
ViTPose++: Vision Transformer for Generic Body Pose EstimationCode3
ZoomNAS: Searching for Whole-body Human Pose Estimation in the WildCode2
Lite Pose: Efficient Architecture Design for 2D Human Pose EstimationCode2
Improving 2D Human Pose Estimation in Rare Camera Views with Synthetic DataCode2
ProbPose: A Probabilistic Approach to 2D Human Pose EstimationCode2
Not All Tokens Are Equal: Human-centric Visual Analysis via Token Clustering TransformerCode2
X-Pose: Detecting Any KeypointsCode2
Metric-Scale Truncation-Robust Heatmaps for 3D Human Pose EstimationCode1
Location-Sensitive Visual Recognition with Cross-IOU LossCode1
Monocular 3D Human Pose Estimation for Sports Broadcasts using Partial Sports Field RegistrationCode1
Keypoint CommunitiesCode1
EvoPose2D: Pushing the Boundaries of 2D Human Pose Estimation using Accelerated Neuroevolution with Weight TransferCode1
Estimating Parkinsonism Severity in Natural Gait Videos of Older Adults with DementiaCode1
Multi-Instance Pose Networks: Rethinking Top-Down Pose EstimationCode1
Learning to Estimate Robust 3D Human Mesh from In-the-Wild Crowded ScenesCode1
EfficientHRNet: Efficient Scaling for Lightweight High-Resolution Multi-Person Pose EstimationCode1
EfficientPose: Scalable single-person pose estimationCode1
AIO-P: Expanding Neural Performance Predictors Beyond Image ClassificationCode1
Show:102550
← PrevPage 1 of 5Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1RTMW-xWB70.2Unverified
2PCNetWB66.4Unverified
3ZoomNAS (V1.0 data)WB65.4Unverified
4RTMPoseWB65.3Unverified
5TCFormerWB64.2Unverified
6ZoomNet (V1.0 data)WB63Unverified
7Sapiens-0.3BWB62Unverified
8ViTPose+-HWB61.2Unverified
9Zauss et al.WB60.4Unverified
10RTMW-mWB58Unverified
#ModelMetricClaimedVerifiedStatus
1UniPoseAP0.76Unverified
2RTMPose-lAP (gt bbox)0.75Unverified
3ED-Pose (R50)AP0.72Unverified
4ViTPose-hAP0.47Unverified
5ViTPose-lAP0.46Unverified
6HRNet-w48AP0.42Unverified
7ViTpose-bAP0.41Unverified
8HRNet-w32AP0.4Unverified
9ViTPose-sAP0.38Unverified
10RTMPose-sAP0.31Unverified
#ModelMetricClaimedVerifiedStatus
1DeciWatchPCK98.8Unverified
2PoseidonPCK97.3Unverified
3SimplePosePCK94.4Unverified
4DKD (ResNet50)PCK94Unverified
5LSTM PMPCK93.6Unverified
#ModelMetricClaimedVerifiedStatus
1SEFDTest AP44.1Unverified
2ResNet-50Test AP30.4Unverified
3Pose2SegTest AP23.8Unverified
#ModelMetricClaimedVerifiedStatus
1mitsimpo10-20% Mask PSNR12Unverified
#ModelMetricClaimedVerifiedStatus
1DA-LLPoseAP5Unverified
#ModelMetricClaimedVerifiedStatus
1DA-LLPoseAP18.6Unverified
#ModelMetricClaimedVerifiedStatus
1DA-LLPoseAP35.6Unverified
#ModelMetricClaimedVerifiedStatus
1DA-LLPoseAP39.1Unverified
#ModelMetricClaimedVerifiedStatus
1DA-LLPoseAP36.2Unverified