Research map / Video / Video generation
DLow: Diversifying Latent Flows for Diverse Human Motion Prediction
Ye Yuan, Kris Kitani
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 reconstruction term E_r (ablation) | DLow · Human3.6M | APD: 13.091 -> 11.741 | Hurt |
| add reconstruction term E_r (ablation) | DLow · Human3.6M | ADE: 0.546 -> 0.425 | Improved |
| add reconstruction term E_r (ablation) | DLow · HumanEva-I | APD: 4.927 -> 4.855 | Hurt |
| add reconstruction term E_r (ablation) | DLow · HumanEva-I | ADE: 0.263 -> 0.251 | Improved |
| add diversity term E_d (ablation) | DLow · Human3.6M | APD: 6.844 -> 11.741 | Improved |
| add diversity term E_d (ablation) | DLow · Human3.6M | ADE: 0.432 -> 0.425 | Improved |
| add diversity term E_d (ablation) | DLow · HumanEva-I | APD: 2.355 -> 4.855 | Improved |
| add diversity term E_d (ablation) | DLow · HumanEva-I | ADE: 0.252 -> 0.251 | Improved |
| add E_d and E_r on top of KL-only training (ablation) | DLow · Human3.6M | APD: 6.383 -> 11.741 | Improved |
| add E_d and E_r on top of KL-only training (ablation) | DLow · Human3.6M | ADE: 0.520 -> 0.425 | Improved |
| add E_d and E_r on top of KL-only training (ablation) | DLow · HumanEva-I | ADE: 0.281 -> 0.251 | Improved |
| replace random CVAE sampling with DLow under controllable energy | CVAE · Human3.6M controllable prediction | Leg Dist: 2.958 -> 1.071 | Improved |
Closest papers by meaning
- Motion Diversification Networks (CVPR 2024 · similarity 0.91)
- Generating Smooth Pose Sequences for Diverse Human Motion Prediction (ICCV 2021 · similarity 0.90)
- BeLFusion: Latent Diffusion for Behavior-Driven Human Motion Prediction (ICCV 2023 · similarity 0.88)
- Deterministic-to-Stochastic Diverse Latent Feature Mapping for Human Motion Synthesis (CVPR 2025 · similarity 0.87)
- Harmonizing Stochasticity and Determinism: Scene-responsive Diverse Human Motion Prediction (NeurIPS 2024 · similarity 0.86)
- MixerMDM: Learnable Composition of Human Motion Diffusion Models (CVPR 2025 · similarity 0.86)
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.
