Research map / Video / Video segmentation and tracking
Temporal Continual Learning with Prior Compensation for Human Motion Prediction
Jianwei Tang, Jiangxin Sun, Xiaotong Lin, lifang zhang, Wei-Shi Zheng, Jian-Fang Hu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add TCL multi-stage training with PCF | PGBIG · Human3.6M | average MPJPE: 66.52 -> 64.97 | Improved |
| add TCL to PGBIG | PGBIG · 3DPW | average MPJPE: 72.49 -> 63.59 | Improved |
| add TCL to PGBIG at 1000ms | PGBIG · 3DPW | MPJPE at 1000ms: 104.1 -> 96.0 | Improved |
| add TCL to PGBIG at 80ms | PGBIG · CMU-MoCap | MPJPE at 80ms: 7.6 -> 7.5 | No clear effect |
| add TCL to PGBIG at 560ms | PGBIG · CMU-MoCap | MPJPE at 560ms: 50.9 -> 48.6 | Improved |
| add TCL to LTD | LTD · Human3.6M | MPJPE at 80ms: 12.7 -> 10.8 | Improved |
| add TCL to Res.Sup. | Res.Sup. · CMU-MoCap | MPJPE at 80ms: 24.4 -> 21.4 | Improved |
| add TCL to STSGCN | STSGCN · AMASS | MPJPE at 1000ms: 69.7 -> 68.1 | Improved |
| add TCL to MotionMixer at 1000ms | MotionMixer · Human3.6M | MPJPE at 1000ms: 117.6 -> 117.1 | No clear effect |
| increase number of tasks from 1 to 3 (ablation) | TCL on PGBIG · Human3.6M | average error: 66.95 -> 65.00 | Improved |
| increase number of tasks from 3 to 8 (ablation) | TCL on PGBIG · Human3.6M | average error: 65.00 -> 65.03 | No clear effect |
| remove PCF (plain multi-stage training) (ablation) | TCL on PGBIG · Human3.6M | task Z1 error after stage S3: 9.30 -> 9.86 | Hurt |
Closest papers by meaning
- Human Motion Prediction via Continual Prior Compensation (TPAMI 2026 · similarity 0.96)
- Continual Predictive Learning From Videos (CVPR 2022 · similarity 0.87)
- Existence Is Chaos: Enhancing 3D Human Motion Prediction with Uncertainty Consideration (AAAI 2024 · similarity 0.84)
- LAL: Enhancing 3D Human Motion Prediction with Latency-aware Auxiliary Learning (CVPR 2025 · similarity 0.84)
- Continual Learning for Motion Prediction Model via Meta-Representation Learning and Optimal Memory Buffer Retention Strategy (CVPR 2024 · similarity 0.84)
- MoML: Online Meta Adaptation for 3D Human Motion Prediction (CVPR 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
