Research map / Video / Video generation
SpaceTimePilot: Generative Rendering of Dynamic Scenes Across Space and Time
Zhening Huang, Hyeonho Jeong, Xuelin Chen, Yulia Gryaditskaya, Tuanfeng Y. Wang, Joan Lasenby, Chun-Hao Huang
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 |
|---|---|---|---|
| compress time embedding with 1D convolution instead of uniform sampling at latent-frame level (ablation) | SpaceTimePilot · Cam×Time test | PSNR: 14.10 -> 14.75 | Improved |
| add static-scene datasets to joint training (ablation) | SpaceTimePilot 1D-Conv · Cam×Time test | PSNR: 14.75 -> 15.41 | Improved |
| add Cam×Time dataset to training (ablation) | SpaceTimePilot 1D-Conv · Cam×Time test | PSNR: 14.75 -> 21.29 | Improved |
| add source-aware camera conditioning (source camera poses) (ablation) | ReCamM+Aug · OpenVideoHD camera control | AbsRot: 11.74 -> 5.63 | Improved |
| add temporal-warping augmentation without source camera conditioning (ablation) | ReCamMaster · OpenVideoHD camera control | AbsRot / RTA30: AbsRot 10.08 -> 11.74; RTA30 10.20% -> 5.93% | Hurt |
| add temporal-warping augmentation without source camera conditioning (ablation) | ReCamMaster · OpenVideoHD camera control | RelRot: RelRot 4.26 -> 3.66; RelTrans 0.32 -> 0.43 | Mixed |
| train with joint static-scene data vs SpaceTimePilot full recipe | ReCamMaster · Cam×Time test (time control) | LPIPS avg: 0.3073 -> 0.2308 | Improved |
| use SpaceTimePilot instead of ReCamMaster | video diffusion camera control · OpenVideoHD VBench | visual quality | No clear effect |
| use MLP compressor instead of 1D-Conv for time signal (ablation) | SpaceTimePilot · qualitative bullet-time | camera smoothness | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- ReCamMaster — ReCamMaster: Camera-Controlled Generative Rendering from A Single Video (ICCV 2025) · better on 1 of 2 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
- Real-time Photorealistic Dynamic Scene Representation and Rendering with 4D Gaussian Splatting (ICLR 2024 · similarity 0.88)
- Spacetime Gaussian Feature Splatting for Real-Time Dynamic View Synthesis (CVPR 2024 · similarity 0.88)
- Diffusion4D: Fast Spatial-temporal Consistent 4D generation via Video Diffusion Models (NeurIPS 2024 · similarity 0.87)
- StateSpaceDiffuser: Bringing Long Context to Diffusion World Models (NeurIPS 2025 · similarity 0.87)
- LongScape: Advancing Long-Horizon Embodied World Models with Context-Aware MoE (NeurIPS 2026 · similarity 0.87)
- ChronoMagic-Bench: A Benchmark for Metamorphic Evaluation of Text-to-Time-lapse Video Generation (NeurIPS 2024 · 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.
