SOTAVerified

Pose Estimation

Pose Estimation is a computer vision task where the goal is to detect the position and orientation of a person or an object. Usually, this is done by predicting the location of specific keypoints like hands, head, elbows, etc. in case of Human Pose Estimation.

A common benchmark for this task is MPII Human Pose

( Image credit: Real-time 2D Multi-Person Pose Estimation on CPU: Lightweight OpenPose )

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 201–250 of 4228 papers

TitleStatusHype
TrackID3x3: A Dataset and Algorithm for Multi-Player Tracking with Identification and Pose Estimation in 3x3 Basketball Full-court VideosCode1
Structure-Aware Correspondence Learning for Relative Pose Estimation—0
Any6D: Model-free 6D Pose Estimation of Novel Objects—0
Selecting and Pruning: A Differentiable Causal Sequentialized State-Space Model for Two-View Correspondence Learning—0
Co-op: Correspondence-based Novel Object Pose Estimation—0
Image as an IMU: Estimating Camera Motion from a Single Motion-Blurred Image—0
Pow3R: Empowering Unconstrained 3D Reconstruction with Camera and Scene Priors—0
Automating 3D Dataset Generation with Neural Radiance FieldsCode0
Probabilistic Prompt Distribution Learning for Animal Pose EstimationCode1
Learning to Efficiently Adapt Foundation Models for Self-Supervised Endoscopic 3D Scene Reconstruction from Any Cameras—0
A Comprehensive Survey on Architectural Advances in Deep CNNs: Challenges, Applications, and Emerging Research Directions—0
EdgeRegNet: Edge Feature-based Multimodal Registration Network between Images and LiDAR Point CloudsCode1
Distilling 3D distinctive local descriptors for 6D pose estimation—0
GIVEPose: Gradual Intra-class Variation Elimination for RGB-based Category-Level Object Pose EstimationCode1
SCJD: Sparse Correlation and Joint Distillation for Efficient 3D Human Pose EstimationCode0
SIR-DIFF: Sparse Image Sets Restoration with Multi-View Diffusion Model—0
Foundation Feature-Driven Online End-Effector Pose Estimation: A Marker-Free and Learning-Free Approach—0
Learning Shape-Independent Transformation via Spherical Representations for Category-Level Object Pose Estimation—0
Uncertainty-Aware Knowledge Distillation for Compact and Efficient 6DoF Pose Estimation—0
STEP: Simultaneous Tracking and Estimation of Pose for Animals and HumansCode0
Bring Your Rear Cameras for Egocentric 3D Human Pose Estimation—0
Online Test-time Adaptation for 3D Human Pose Estimation: A Practical Perspective with Estimated 2D Poses—0
Fast and Robust Localization for Humanoid Soccer Robot via Iterative Landmark Matching—0
Clothes-Changing Person Re-identification Based On Skeleton Dynamics—0
Consistent multi-animal pose estimation in cattle using dynamic Kalman filter based trackingCode0
VicaSplat: A Single Run is All You Need for 3D Gaussian Splatting and Camera Estimation from Unposed Video Frames—0
6D Object Pose Tracking in Internet Videos for Robotic Manipulation—0
Better Together: Unified Motion Capture and 3D Avatar Reconstruction—0
GenHPE: Generative Counterfactuals for 3D Human Pose Estimation with Radio Frequency Signals—0
Keypoint Detection and Description for Raw Bayer Images—0
Keypoint Semantic Integration for Improved Feature Matching in Outdoor Agricultural Environments—0
Acoustic Neural 3D Reconstruction Under Pose Drift—0
Endo-FASt3r: Endoscopic Foundation model Adaptation for Structure from motion—0
Multi-Modal 3D Mesh Reconstruction from Images and Text—0
HumanMM: Global Human Motion Recovery from Multi-shot VideosCode2
AthletePose3D: A Benchmark Dataset for 3D Human Pose Estimation and Kinematic Validation in Athletic MovementsCode2
PoseLess: Depth-Free Vision-to-Joint Control via Direct Image Mapping with VLM—0
AxisPose: Model-Free Matching-Free Single-Shot 6D Object Pose Estimation via Axis Generation—0
NeuraLoc: Visual Localization in Neural Implicit Map with Dual Complementary Features—0
Fish2Mesh Transformer: 3D Human Mesh Recovery from Egocentric Vision—0
Multi-Grained Feature Pruning for Video-Based Human Pose Estimation—0
SplatPose: Geometry-Aware 6-DoF Pose Estimation from Single RGB Image via 3D Gaussian Splatting—0
Novel Object 6D Pose Estimation with a Single Reference ViewCode1
GaussianCAD: Robust Self-Supervised CAD Reconstruction from Three Orthographic Views Using 3D Gaussian Splatting—0
A Study on the Matching Rate of Dance Movements Using 2D Skeleton Detection and 3D Pose Estimation: Why Is SEVENTEEN's Performance So Bita-Zoroi (Perfectly Synchronized)?—0
ReynoldsFlow: Exquisite Flow Estimation via Reynolds Transport TheoremCode0
MarsLGPR: Mars Rover Localization with Ground Penetrating Radar—0
Machine Learning in Biomechanics: Key Applications and Limitations in Walking, Running, and Sports Movements—0
Improving 6D Object Pose Estimation of metallic Household and Industry Objects—0
Active 6D Pose Estimation for Textureless Objects using Multi-View RGB Frames—0
Show:102550
← PrevPage 5 of 85Next →
All datasetsCOCO test-devMPII Human PoseOCHumanLeeds Sports PosesCrowdPoseCOCO val2017AICCOCO (Common Objects in Context)InLocITOP front-viewJ-HMDBMPII Single Person

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1yoloposeAP5090.3—Unverified
2ViTPose (ViTAE-G, ensemble)AP81.1—Unverified
3ViTPose (ViTAE-G)AP80.9—Unverified
4UDP-Pose-PSA(384x288)AP79.5—Unverified
5PoseBH-HAP79.5—Unverified
64xRSN-50 (ensemble)AP79.2—Unverified
7UDP-Pose-PSA(256x192)AP78.9—Unverified
8CCM+AP78.9—Unverified
94xRSN-50AP78.6—Unverified
10PCT (256x256)AP78.3—Unverified
#ModelMetricClaimedVerifiedStatus
1PCT (swin-l, test set)PCKh-0.594.3—Unverified
2Soft-gated Skip ConnectionsPCKh-0.594.1—Unverified
3Cascade Feature AggregationPCKh-0.593.9—Unverified
4PCT (swin-b, test set)PCKh-0.593.8—Unverified
5TransPosePCKh-0.593.5—Unverified
6UniHCP (FT)PCKh-0.593.2—Unverified
74xRSN-50PCKh-0.593—Unverified
8UniPosePCKh-0.592.7—Unverified
9MSPNPCKh-0.592.6—Unverified
10Spatial ContextPCKh-0.592.5—Unverified
#ModelMetricClaimedVerifiedStatus
1ViTPose (ViTAE-G, GT bounding boxes)Test AP93.3—Unverified
2UniHCP (direct eval)Test AP87.4—Unverified
3PoseBH-HTest AP87—Unverified
4RTMPose(RTMPose-l, GT bounding boxes)Test AP80.3—Unverified
5TransPose-HValidation AP62.3—Unverified
6BBox-Mask-Pose 2xTest AP48.3—Unverified
7BUCTD (CID-W32)Test AP47.2—Unverified
8HQNet (ViT-L)Test AP45.6—Unverified
9CID (HRNet-W48)Test AP45—Unverified
10MaskPose-bTest AP45—Unverified
#ModelMetricClaimedVerifiedStatus
1OmniPosePCK99.5—Unverified
2Soft-gated Skip ConnectionsPCK94.8—Unverified
3UniPosePCK94.5—Unverified
4Residual Hourglass + ASR + AHOPCK94.5—Unverified
5Chou et al. arXiv'17PCK94—Unverified
6Pyramid Residual Modules (PRMs)PCK93.9—Unverified
7Stacked hourglass + Inception-resnetPCK93.9—Unverified
8Multi-Context AttentionPCK92.6—Unverified
9FPDPCK90.8—Unverified
10Part heatmap regression (ResNet-152)PCK90.7—Unverified
#ModelMetricClaimedVerifiedStatus
1BUCTD-W48 (w/cond. input from PETR, and generative sampling)AP78.5—Unverified
2ViTPose-GAP78.3—Unverified
3BUCTD-W48 (w/cond. input from PETR)AP76.7—Unverified
4SwinV2-L 1K-MIMAP75.5—Unverified
5SwinV2-B 1K-MIMAP74.9—Unverified
6BUCTD-W48AP72.9—Unverified
7OpenPifPafAP70.5—Unverified
8MIPNet (HRNet-W48)AP70—Unverified
9KAPAO-LAP68.9—Unverified
10KAPAO-MAP67.1—Unverified
#ModelMetricClaimedVerifiedStatus
1CCNet (ViTPose-B_GT-bbox_256x192)AP78.1—Unverified
2MogaNet-B (384x288)AP77.3—Unverified
3ViTPose-B (Single-task_GT-bbox_256x192)AP77.3—Unverified
4MogaNet-S (384x288)AP76.4—Unverified
5Bias (HRNet_256x192)AP75.8—Unverified
6ViTPose-B (Single-task_Det-bbox_256x192)AP75.8—Unverified
7HRNet (256x192)AP75.3—Unverified
8MogaNet-S (256x192)AP74.9—Unverified
9MogaNet-T (256x192)AP73.2—Unverified
10RLE (256x192)AP71.3—Unverified
#ModelMetricClaimedVerifiedStatus
1Hulk(Finetune, ViT-L)AP37.1—Unverified
2Hulk(Finetune, ViT-B)AP35.6—Unverified
3HRFormer (HRFomer-B)AP34.4—Unverified
4UniHCP (finetune)AP33.6—Unverified
5HRNet (HRNet-w48 )AP33.5—Unverified
6HRNet (HRNet-w32)AP32.3—Unverified
7HRFormer (HRFomer-S)AP31.6—Unverified
8SimpleBaseline (ResNet-152)AP29.9—Unverified
9SimpleBaseline (ResNet-101)AP29.4—Unverified
10SimpleBaseline (ResNet-50)AP28—Unverified
#ModelMetricClaimedVerifiedStatus
1BUCTD (PETR, with generative sampling)APL83.7—Unverified
2OmniPose (WASPv2)AP79.5—Unverified
3MetaPrompt-SDAP79—Unverified
4Hulk(Finetune, ViT-L)AP78.7—Unverified
5BUCTD (PETR, with generative sampling)AP77.8—Unverified
6Hulk(Finetune, ViT-B)AP77.5—Unverified
7I²R-Net (1st stage:HRFormer-B)AP77.3—Unverified
8PATH (Partial FT)AP77.1—Unverified
9SOLIDER (swin-B)AP76.6—Unverified
10PEFORMER-Xcit-dino-p8AP72.6—Unverified
#ModelMetricClaimedVerifiedStatus
1GIM-DKMDUC1-Acc@0.25m,10°57.1—Unverified
2GIM-LoFTRDUC1-Acc@0.25m,10°54.5—Unverified
3GIM-SuperGlueDUC1-Acc@0.25m,10°53.5—Unverified
4DKMDUC1-Acc@0.25m,10°51.5—Unverified
5SuperGlueDUC1-Acc@0.25m,10°49—Unverified
6LoFTRDUC1-Acc@0.25m,10°47.5—Unverified
#ModelMetricClaimedVerifiedStatus
1AdaPoseMean mAP93.38—Unverified
2DECA-D3Mean mAP88.75—Unverified
3V2V-PoseNetMean mAP88.74—Unverified
4A2JMean mAP88—Unverified
5RENMean mAP84.9—Unverified
6Multi-task learning + viewpoint invarianceMean mAP77.4—Unverified
#ModelMetricClaimedVerifiedStatus
1SimpleBaseline + HANetMean PCK@0.299.6—Unverified
2DeciWatchMean PCK@0.299—Unverified
3LSTM PMMean PCK@0.293.6—Unverified
4CPMMean PCK@0.291.9—Unverified
5UniTrack_i18Mean PCK@0.280.5—Unverified
#ModelMetricClaimedVerifiedStatus
14xRSN-50PCKh@0.593—Unverified
2RefinePCKh@0.592.1—Unverified
3EfficientPose IVPCKh@0.591.2—Unverified
4OpenPosePCKh@0.588.8—Unverified
5Adversarial LearningPCKh@0.588.6—Unverified
#ModelMetricClaimedVerifiedStatus
1OmniPoseMean PCK@0.299.4—Unverified
2UniPose-LSTMMean PCK@0.299.3—Unverified
3LSTM PMMean PCK@0.297.7—Unverified
4Thin-SlicingMean PCK@0.296.5—Unverified
5Iqbal et al.Mean PCK@0.281.1—Unverified
#ModelMetricClaimedVerifiedStatus
1DP-RCNN-DeepLab (ResNet-101)AP68—Unverified