Human Pose Forecasting On Human36M
Metrics
Average MPJPE (mm) @ 1000 ms
Average MPJPE (mm) @ 400ms
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | Average MPJPE (mm) @ 1000 ms | Average MPJPE (mm) @ 400ms |
|---|---|---|
| a-unified-masked-autoencoder-with-patchified | 112.1 | 61.6 |
| a-mixer-layer-is-worth-one-graph-convolution | 108.6 | 56.7 |
| accurate-and-diverse-sampling-of-sequences | - | - |
| diverse-human-motion-prediction-guided-by | - | - |
| generating-smooth-pose-sequences-for-diverse | - | - |
| multiscale-residual-learning-of-graph | 81.95 | 56.41 |
| motionmixer-mlp-based-3d-human-body-pose | 111.0 | 59.3 |
| a-generic-diffusion-based-approach-for-3d | - | - |
| eqmotion-equivariant-multi-agent-motion | 106.9 | 55.0 |
| msr-gcn-multi-scale-residual-graph | 114.2 | 62.9 |
| hp-gan-probabilistic-3d-human-motion | - | - |
| gcnext-towards-the-unity-of-graph | 64.7 | 30.5 |
| towards-accurate-human-motion-prediction-via | 109.2 | 55.5 |
| space-time-separable-graph-convolutional-1 | 117.0 | 65.8 |
| action-agnostic-human-pose-forecasting | - | - |
| imitation-learning-for-human-pose-prediction | - | - |
| auxiliary-tasks-benefit-3d-skeleton-based | 107 | 54.1 |
| the-pose-knows-video-forecasting-by | - | - |
| pages-iai-uni-bonn-de-https-pages-iai-uni | 149.2 | 80.8 |
| mt-vae-learning-motion-transformations-to | - | - |
| multi-graph-convolution-network-for-pose | 72.9 | 34.4 |
| anypose-anytime-3d-human-pose-forecasting-via | 128.2 | 80.6 |
| back-to-mlp-a-simple-baseline-for-human | 109.4 | 57.3 |
| remembering-what-is-important-a-factorised | - | 34.8 |
| deep-unsupervised-clustering-with-gaussian | - | - |
| structural-rnn-deep-learning-on-spatio | - | - |
| transfusion-a-practical-and-effective | - | - |
| learning-trajectory-dependencies-for-human | 113.0 | 63.5 |
| diverse-trajectory-forecasting-with | - | - |
| recurrent-network-models-for-human-dynamics | - | - |
| belfusion-latent-diffusion-for-behavior | - | - |
| progressively-generating-better-initial | 110.3 | 58.5 |
| diverse-human-motion-prediction-via-gumbel | - | - |