Multi Person Pose Estimation On Crowdpose
평가 지표
AP Easy
AP Hard
AP Medium
mAP @0.5:0.95
평가 결과
이 벤치마크에서 각 모델의 성능 결과
비교 표
| 모델 이름 | AP Easy | AP Hard | AP Medium | mAP @0.5:0.95 |
|---|---|---|---|---|
| mask-r-cnn | 69.4 | 45.8 | 57.9 | 57.2 |
| transpose-towards-explainable-human-pose | 79.5 | 62.2 | 72.9 | 71.8 |
| rmpe-regional-multi-person-pose-estimation | 71.2 | 51.1 | 61.4 | 61.0 |
| rtmo-towards-high-performance-one-stage-real | 88.8 | 77.2 | 84.7 | 83.8 |
| self-constrained-inference-optimization-on | - | - | 72.2 | 71.5 |
| openpose-realtime-multi-person-2d-pose | 62.7 | 32.3 | 58.7 | - |
| centerhmr-a-bottom-up-single-shot-method-for | - | - | - | 58.6 |
| greedy-offset-guided-keypoint-grouping-for | 73.8 | 54.8 | 66.2 | 65.2 |
| bottom-up-higher-resolution-networks-for | 75.8 | 58.9 | 68.1 | 67.6 |
| rethinking-pose-estimation-in-crowds | 83.9 | 72.3 | 79.0 | 78.5 |
| scalenas-one-shot-learning-of-scale-aware | - | - | - | 71.3 |
| lite-pose-efficient-architecture-design-for | - | - | - | 58.3 |
| centerhmr-a-bottom-up-single-shot-method-for | - | - | - | 55.6 |
| bapose-bottom-up-pose-estimation-with | 79.9 | 61.3 | 73.4 | 72.2 |
| multi-hypothesis-pose-networks-rethinking-top | 78.1 | 59.4 | 71.1 | 70.0 |
| the-center-of-attention-center-keypoint-1 | 76.6 | 61.5 | 70.0 | 69.4 |
| crowdpose-efficient-crowded-scenes-pose | 75.5 | 57.4 | 66.3 | 66.0 |
| i-2r-net-intra-and-inter-human-relation | 83.8 | 69.3 | 78.1 | 77.4 |
| hrformer-high-resolution-transformer-for | 80.0 | 62.4 | 73.5 | 72.4 |
| single-stage-multi-person-pose-machines | 70.3 | 55.7 | 64.5 | 63.7 |
| simple-baselines-for-human-pose-estimation | 71.4 | 51.2 | 61.2 | 60.8 |
| explicit-box-detection-unifies-end-to-end | 83.0 | 68.3 | 77.3 | 76.6 |
| human-pose-estimation-for-real-world-crowded | 75.2 | 53.1 | 66.6 | 65.5 |