Pose Estimation On Coco Test Dev
评估指标
AP
AP50
AP75
APL
APM
AR
评测结果
各个模型在此基准测试上的表现结果
比较表格
| 模型名称 | AP | AP50 | AP75 | APL | APM | AR |
|---|---|---|---|---|---|---|
| simple-baselines-for-human-pose-estimation | 73.7 | 91.9 | 81.1 | 80 | 70.3 | 79 |
| mask-r-cnn | 63.1 | 87.3 | 68.7 | 71.4 | - | - |
| deep-multi-task-networks-for-occluded | 75.7 | 90.3 | 76.3 | 79.5 | 80.7 | - |
| omnipose-a-multi-scale-framework-for-multi | 76.4 | 92.6 | 83.7 | 82.6 | 72.6 | 81.2 |
| revisiting-unreasonable-effectiveness-of-data | 64.4 | 85.7 | 70.7 | 69.8 | 61.8 | - |
| lite-hrnet-a-lightweight-high-resolution | 69.7 | 90.7 | 77.5 | 75.0 | 66.9 | 75.4 |
| multi-hypothesis-pose-networks-rethinking-top | 75.7 | 92.4 | 83.3 | 81.2 | 71.4 | 80.5 |
| 2103-15320 | 72.2 | 90.9 | 80.1 | 78.8 | 69.1 | - |
| posefix-model-agnostic-general-human-pose | 74.7 | 91.2 | 81.9 | 81.2 | 71.1 | 79.9 |
| rethinking-on-multi-stage-networks-for-human | 76.1 | 93.4 | 83.8 | 81.5 | 72.3 | 81.6 |
| transpose-towards-explainable-human-pose | 75 | 92.2 | 82.3 | 81.1 | 71.3 | - |
| the-devil-is-in-the-details-delving-into | 76.5 | 92.7 | 84 | 73.0 | 82.4 | 81.6 |
| distribution-aware-coordinate-representation | 77.4 | 92.6 | 84.6 | 83.7 | 73.6 | 82.3 |
| realtime-multi-person-2d-pose-estimation | 61.8 | 84.9 | 67.5 | 68.2 | 57.1 | 66.5 |
| vitpose-simple-vision-transformer-baselines | 81.1 | 95.0 | 88.2 | 86.0 | 77.8 | 85.6 |
| vitpose-simple-vision-transformer-baselines | 80.9 | 94.8 | 88.1 | 85.9 | 77.5 | 85.4 |
| rethinking-keypoint-representations-modeling | 70.3 | 91.2 | 77.8 | 76.8 | 66.3 | 77.7 |
| rmpe-regional-multi-person-pose-estimation | 72.3 | 89.2 | 79.1 | 78.6 | 68.0 | - |
| cascaded-pyramid-network-for-multi-person | 72.1 | 91.4 | 80.0 | 77.2 | - | 78.5 |
| yolo-pose-enhancing-yolo-for-multi-person | - | 90.3 | - | - | - | - |
| simple-pose-rethinking-and-improving-a-bottom | 68.1 | - | - | 70.5 | 66.8 | 88.2 |
| revealing-the-dark-secrets-of-masked-image | 77.2 | - | - | - | - | - |
| human-pose-as-compositional-tokens | 78.3 | 92.9 | 85.9 | - | - | - |
| on-the-calibration-of-human-pose-estimation | - | - | - | - | - | - |
| learning-delicate-local-representations-for | 78.6 | 94.3 | 86.6 | 75.5 | 83.3 | 83.8 |
| vipnas-efficient-video-pose-estimation-via | 70.3 | 90.7 | 78.8 | 75.5 | 67.3 | 77.3 |
| deep-high-resolution-representation-learning | 77 | 92.7 | 84.5 | 83.1 | 73.4 | 82 |
| lite-hrnet-a-lightweight-high-resolution | 66.9 | 89.4 | 74.4 | 72.2 | 64.0 | 72.6 |
| learning-delicate-local-representations-for | 79.2 | 94.4 | 87.1 | 76.1 | 83.8 | 84.1 |
| hrformer-high-resolution-transformer-for | 76.2 | 92.7 | 83.8 | 82.3 | 72.5 | 81.2 |
| dite-hrnet-dynamic-lightweight-high | 70.6 | 90.8 | 78.2 | 76.1 | 67.4 | 76.4 |
| 模型 32 | 70.8 | - | - | - | - | - |
| dpit-dual-pipeline-integrated-transformer-for | 74.6 | 91.9 | 82.1 | 80.6 | 71.3 | 79.9 |
| vipnas-efficient-video-pose-estimation-via | 73.9 | 91.7 | 82 | 79.5 | 70.5 | 80.4 |
| revealing-the-dark-secrets-of-masked-image | 76.7 | - | - | - | - | - |
| rethinking-keypoint-representations-modeling | 63.8 | 88.4 | 70.4 | 71.7 | 58.6 | 71.2 |
| rethinking-keypoint-representations-modeling | 68.8 | 90.5 | 76.5 | 76 | 64.3 | 76.3 |
| towards-accurate-multi-person-pose-estimation | 64.9 | 85.5 | 71.3 | 70.0 | - | 69.7 |
| towards-high-performance-human-keypoint | 78.9 | 93.8 | 86 | 84.5 | 75 | 83.6 |
| rmpe-regional-multi-person-pose-estimation | 61.8 | 83.7 | 69.8 | 67.6 | 58.6 | - |
| openpose-realtime-multi-person-2d-pose | 64.2 | 86.2 | 70.1 | 68.8 | 61 | - |
| evopose2d-pushing-the-boundaries-of-2d-human | 76.8 | 92.5 | 84.3 | 82.5 | 73.5 | 81.7 |
| polarized-self-attention-towards-high-quality-1 | 79.5 | 93.6 | 85.9 | 84.3 | 76.3 | 81.9 |
| polarized-self-attention-towards-high-quality-1 | 78.9 | 93.6 | 85.8 | 83.6 | 76.1 | 81.4 |
| directpose-direct-end-to-end-multi-person | 63.3 | 86.7 | 69.4 | 71.2 | 57.8 | - |
| cascaded-pyramid-network-for-multi-person | 73.0 | 91.7 | 80.9 | 78.1 | - | 79.0 |