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
Showing 101–125 of 4228 papers
All datasetsCOCO test-devMPII Human PoseOCHumanLeeds Sports PosesCrowdPoseCOCO val2017AICCOCO (Common Objects in Context)InLocITOP front-viewJ-HMDBMPII Single Person
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | yolopose | AP50 | 90.3 | — | Unverified |
| 2 | ViTPose (ViTAE-G, ensemble) | AP | 81.1 | — | Unverified |
| 3 | ViTPose (ViTAE-G) | AP | 80.9 | — | Unverified |
| 4 | PoseBH-H | AP | 79.5 | — | Unverified |
| 5 | UDP-Pose-PSA(384x288) | AP | 79.5 | — | Unverified |
| 6 | 4xRSN-50 (ensemble) | AP | 79.2 | — | Unverified |
| 7 | UDP-Pose-PSA(256x192) | AP | 78.9 | — | Unverified |
| 8 | CCM+ | AP | 78.9 | — | Unverified |
| 9 | 4xRSN-50 | AP | 78.6 | — | Unverified |
| 10 | PCT (256x256) | AP | 78.3 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | PCT (swin-l, test set) | PCKh-0.5 | 94.3 | — | Unverified |
| 2 | Soft-gated Skip Connections | PCKh-0.5 | 94.1 | — | Unverified |
| 3 | Cascade Feature Aggregation | PCKh-0.5 | 93.9 | — | Unverified |
| 4 | PCT (swin-b, test set) | PCKh-0.5 | 93.8 | — | Unverified |
| 5 | TransPose | PCKh-0.5 | 93.5 | — | Unverified |
| 6 | UniHCP (FT) | PCKh-0.5 | 93.2 | — | Unverified |
| 7 | 4xRSN-50 | PCKh-0.5 | 93 | — | Unverified |
| 8 | UniPose | PCKh-0.5 | 92.7 | — | Unverified |
| 9 | MSPN | PCKh-0.5 | 92.6 | — | Unverified |
| 10 | Spatial Context | PCKh-0.5 | 92.5 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | ViTPose (ViTAE-G, GT bounding boxes) | Test AP | 93.3 | — | Unverified |
| 2 | UniHCP (direct eval) | Test AP | 87.4 | — | Unverified |
| 3 | PoseBH-H | Test AP | 87 | — | Unverified |
| 4 | RTMPose(RTMPose-l, GT bounding boxes) | Test AP | 80.3 | — | Unverified |
| 5 | TransPose-H | Validation AP | 62.3 | — | Unverified |
| 6 | BBox-Mask-Pose 2x | Test AP | 48.3 | — | Unverified |
| 7 | BUCTD (CID-W32) | Test AP | 47.2 | — | Unverified |
| 8 | HQNet (ViT-L) | Test AP | 45.6 | — | Unverified |
| 9 | MaskPose-b | Test AP | 45 | — | Unverified |
| 10 | CID (HRNet-W48) | Test AP | 45 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | OmniPose | PCK | 99.5 | — | Unverified |
| 2 | Soft-gated Skip Connections | PCK | 94.8 | — | Unverified |
| 3 | Residual Hourglass + ASR + AHO | PCK | 94.5 | — | Unverified |
| 4 | UniPose | PCK | 94.5 | — | Unverified |
| 5 | Chou et al. arXiv'17 | PCK | 94 | — | Unverified |
| 6 | Pyramid Residual Modules (PRMs) | PCK | 93.9 | — | Unverified |
| 7 | Stacked hourglass + Inception-resnet | PCK | 93.9 | — | Unverified |
| 8 | Multi-Context Attention | PCK | 92.6 | — | Unverified |
| 9 | FPD | PCK | 90.8 | — | Unverified |
| 10 | Part heatmap regression (ResNet-152) | PCK | 90.7 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | BUCTD-W48 (w/cond. input from PETR, and generative sampling) | AP | 78.5 | — | Unverified |
| 2 | ViTPose-G | AP | 78.3 | — | Unverified |
| 3 | BUCTD-W48 (w/cond. input from PETR) | AP | 76.7 | — | Unverified |
| 4 | SwinV2-L 1K-MIM | AP | 75.5 | — | Unverified |
| 5 | SwinV2-B 1K-MIM | AP | 74.9 | — | Unverified |
| 6 | BUCTD-W48 | AP | 72.9 | — | Unverified |
| 7 | OpenPifPaf | AP | 70.5 | — | Unverified |
| 8 | MIPNet (HRNet-W48) | AP | 70 | — | Unverified |
| 9 | KAPAO-L | AP | 68.9 | — | Unverified |
| 10 | KAPAO-M | AP | 67.1 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | CCNet (ViTPose-B_GT-bbox_256x192) | AP | 78.1 | — | Unverified |
| 2 | MogaNet-B (384x288) | AP | 77.3 | — | Unverified |
| 3 | ViTPose-B (Single-task_GT-bbox_256x192) | AP | 77.3 | — | Unverified |
| 4 | MogaNet-S (384x288) | AP | 76.4 | — | Unverified |
| 5 | Bias (HRNet_256x192) | AP | 75.8 | — | Unverified |
| 6 | ViTPose-B (Single-task_Det-bbox_256x192) | AP | 75.8 | — | Unverified |
| 7 | HRNet (256x192) | AP | 75.3 | — | Unverified |
| 8 | MogaNet-S (256x192) | AP | 74.9 | — | Unverified |
| 9 | MogaNet-T (256x192) | AP | 73.2 | — | Unverified |
| 10 | RLE (256x192) | AP | 71.3 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | Hulk(Finetune, ViT-L) | AP | 37.1 | — | Unverified |
| 2 | Hulk(Finetune, ViT-B) | AP | 35.6 | — | Unverified |
| 3 | HRFormer (HRFomer-B) | AP | 34.4 | — | Unverified |
| 4 | UniHCP (finetune) | AP | 33.6 | — | Unverified |
| 5 | HRNet (HRNet-w48 ) | AP | 33.5 | — | Unverified |
| 6 | HRNet (HRNet-w32) | AP | 32.3 | — | Unverified |
| 7 | HRFormer (HRFomer-S) | AP | 31.6 | — | Unverified |
| 8 | SimpleBaseline (ResNet-152) | AP | 29.9 | — | Unverified |
| 9 | SimpleBaseline (ResNet-101) | AP | 29.4 | — | Unverified |
| 10 | SimpleBaseline (ResNet-50) | AP | 28 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | BUCTD (PETR, with generative sampling) | APL | 83.7 | — | Unverified |
| 2 | OmniPose (WASPv2) | AP | 79.5 | — | Unverified |
| 3 | MetaPrompt-SD | AP | 79 | — | Unverified |
| 4 | Hulk(Finetune, ViT-L) | AP | 78.7 | — | Unverified |
| 5 | BUCTD (PETR, with generative sampling) | AP | 77.8 | — | Unverified |
| 6 | Hulk(Finetune, ViT-B) | AP | 77.5 | — | Unverified |
| 7 | I²R-Net (1st stage:HRFormer-B) | AP | 77.3 | — | Unverified |
| 8 | PATH (Partial FT) | AP | 77.1 | — | Unverified |
| 9 | SOLIDER (swin-B) | AP | 76.6 | — | Unverified |
| 10 | PEFORMER-Xcit-dino-p8 | AP | 72.6 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | GIM-DKM | [email protected],10° | 57.1 | — | Unverified |
| 2 | GIM-LoFTR | [email protected],10° | 54.5 | — | Unverified |
| 3 | GIM-SuperGlue | [email protected],10° | 53.5 | — | Unverified |
| 4 | DKM | [email protected],10° | 51.5 | — | Unverified |
| 5 | SuperGlue | [email protected],10° | 49 | — | Unverified |
| 6 | LoFTR | [email protected],10° | 47.5 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | AdaPose | Mean mAP | 93.38 | — | Unverified |
| 2 | DECA-D3 | Mean mAP | 88.75 | — | Unverified |
| 3 | V2V-PoseNet | Mean mAP | 88.74 | — | Unverified |
| 4 | A2J | Mean mAP | 88 | — | Unverified |
| 5 | REN | Mean mAP | 84.9 | — | Unverified |
| 6 | Multi-task learning + viewpoint invariance | Mean mAP | 77.4 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | SimpleBaseline + HANet | Mean [email protected] | 99.6 | — | Unverified |
| 2 | DeciWatch | Mean [email protected] | 99 | — | Unverified |
| 3 | LSTM PM | Mean [email protected] | 93.6 | — | Unverified |
| 4 | CPM | Mean [email protected] | 91.9 | — | Unverified |
| 5 | UniTrack_i18 | Mean [email protected] | 80.5 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | 4xRSN-50 | [email protected] | 93 | — | Unverified |
| 2 | Refine | [email protected] | 92.1 | — | Unverified |
| 3 | EfficientPose IV | [email protected] | 91.2 | — | Unverified |
| 4 | OpenPose | [email protected] | 88.8 | — | Unverified |
| 5 | Adversarial Learning | [email protected] | 88.6 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | OmniPose | Mean [email protected] | 99.4 | — | Unverified |
| 2 | UniPose-LSTM | Mean [email protected] | 99.3 | — | Unverified |
| 3 | LSTM PM | Mean [email protected] | 97.7 | — | Unverified |
| 4 | Thin-Slicing | Mean [email protected] | 96.5 | — | Unverified |
| 5 | Iqbal et al. | Mean [email protected] | 81.1 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | DP-RCNN-DeepLab (ResNet-101) | AP | 68 | — | Unverified |