Research map / Video / Video generation
Pulp Motion: Framing-aware multimodal camera and human motion generation
Robin Courant, Xi WANG, David Loiseaux, Marc Christie, Vicky Kalogeiton
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 auxiliary sampling (w_z=0.25) to dual-modality generation (ablation) | (x,y) DiT · PulpMotion mixed subset | FD_framing: 4.90 -> 3.37 | Improved |
| add auxiliary sampling to dual-modality generation (ablation) | (x,y) DiT · PulpMotion mixed subset | Out-rate (%): 25.98 -> 16.76 | Improved |
| add auxiliary sampling to dual-modality generation (ablation) | (x,y) MAR · PulpMotion mixed subset | FD_framing: 8.51 -> 6.42 | Improved |
| add auxiliary sampling to independent generation (ablation) | (x)(y) DiT · PulpMotion mixed subset | FD_framing: 11.21 -> 8.24 | Improved |
| add auxiliary sampling (ablation) | (x,y) DiT · PulpMotion mixed subset | human TMR-Score: 23.50 -> 25.05 | Improved |
| add auxiliary sampling (ablation) | (x,y) MAR · PulpMotion mixed subset | camera CLaTr-Score: 42.84 -> 45.96 | Improved |
| add auxiliary sampling (ablation) | (x,y) DiT · PulpMotion mixed subset | FD_TMR: 372.61 -> 431.54 | Hurt |
| condition camera on generated human motion only instead of joint generation (ablation) | (x)+Director DiT · PulpMotion mixed subset | FD_framing: 22.21 vs 8.24 for (x)(y)+Aux | Hurt |
| add auxiliary modality z as a third generated modality without Aux sampling (ablation) | (x,y,z) DiT · PulpMotion mixed subset | FD_framing: 4.90 -> 4.18 | Improved |
| add auxiliary modality z as a third generated modality without Aux sampling (ablation) | (x,y,z) DiT · PulpMotion mixed subset | camera CLaTr-Score: 30.75 -> 23.34 | Hurt |
| replace ReDi representation guidance with auxiliary sampling (ablation) | (x,y) DiT · PulpMotion mixed subset | FD_framing: 5.57 -> 3.37 | Improved |
| increase auxiliary guidance weight w_z beyond the moderate value (ablation) | (x,y)+Aux DiT · PulpMotion mixed subset | FD_TMR: 431.54 -> 493.53 | Hurt |
Closest papers by meaning
- Auteur: Language-Driven Cinematographic Framing for Human-Centric Video Generation (NeurIPS 2026 · similarity 0.91)
- Unveiling the Value of Motion for Cinematic Camera Trajectories (NeurIPS 2026 · similarity 0.90)
- MotionCharacter: Fine-Grained Motion Controllable Human Video Generation (AAAI 2026 · similarity 0.89)
- CineTechBench: A Benchmark for Cinematographic Technique Understanding and Generation (NeurIPS 2025 · similarity 0.87)
- AMD: Autoregressive Motion Diffusion (AAAI 2024 · similarity 0.87)
- Towards Understanding Camera Motions in Any Video (NeurIPS 2025 · similarity 0.87)
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.
