Research map / Label-efficient and robust learning / Knowledge distillation
Multi-Objective Diverse Human Motion Prediction With Knowledge Distillation
Hengbo Ma, Jiachen Li, Ramtin Hosseini, Masayoshi Tomizuka, Chiho Choi
Check 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 |
|---|---|---|---|
| Use the multi-objective predictor with accuracy and diversity priors | Human motion prediction · Human3.6M | ADE: 0.425 -> 0.414 | Improved |
| Use the multi-objective predictor with accuracy and diversity priors | Human motion prediction · Human3.6M | APD: 11.74 -> 14.24 | Improved |
| Use the multi-objective predictor with accuracy and diversity priors | Human motion prediction · HumanEva-I | ADE: 0.251 -> 0.228 | Improved |
| Use the multi-objective predictor with accuracy and diversity priors | Human motion prediction · HumanEva-I | APD: 4.855 -> 5.786 | Improved |
| Use the proposed predictor | Human motion prediction · Human3.6M | FDE: 0.518 -> 0.516 | Improved |
| Use the proposed predictor | Human motion prediction · HumanEva-I | FDE: 0.268 -> 0.236 | Improved |
| Add a short-term oracle with τ=25 instead of using a full-horizon oracle with τ=100 (ablation) | Human3.6M predictor · Human3.6M | APD: 5.651 -> 5.927 | Improved |
| Add a short-term oracle with τ=25 instead of using a full-horizon oracle with τ=100 (ablation) | Human3.6M predictor · Human3.6M | ADE: 0.402 -> 0.404 | Mixed |
| Add a short-term oracle with τ=25 instead of using a full-horizon oracle with τ=100 (ablation) | Human3.6M predictor · Human3.6M | FDE: 0.497 -> 0.501 | Mixed |
| Increase the number of accuracy-prior samples during testing (ablation) | Proposed predictor · Human3.6M | APD: 18.79 -> 5.927 | Hurt |
| Increase the number of accuracy-prior samples during testing (ablation) | Proposed predictor · Human3.6M | ADE: 0.941 -> 0.404 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Dlow — DLOW: Domain Flow for Adaptation and Generalization (CVPR 2019) · better on 6 of 6 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Motion Diversification Networks (CVPR 2024 · similarity 0.88)
- SimpliHuMoN: Simplifying Human Motion Prediction (NeurIPS 2026 · similarity 0.87)
- Harmonizing Stochasticity and Determinism: Scene-responsive Diverse Human Motion Prediction (NeurIPS 2024 · similarity 0.87)
- Existence Is Chaos: Enhancing 3D Human Motion Prediction with Uncertainty Consideration (AAAI 2024 · similarity 0.86)
- MoFlow: One-Step Flow Matching for Human Trajectory Forecasting via Implicit Maximum Likelihood Estimation based Distillation (CVPR 2025 · similarity 0.85)
- LAL: Enhancing 3D Human Motion Prediction with Latency-aware Auxiliary Learning (CVPR 2025 · similarity 0.85)
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.
