Research map / 3D vision / 3D generation
FVGen: Accelerating Novel-View Synthesis with Adversarial Video Diffusion Distillation
Wenbin Teng, Gonglin Chen, Haiwei Chen, Yajie Zhao
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 GAN-based student initialization (ablation) | FVGen · ablation study | PSNR: 8.62 -> 17.50 | Improved |
| add GAN-based student initialization (ablation) | FVGen · ablation study | FID: 40.17 -> 20.54 | Improved |
| add DMD distribution matching after GAN (ablation) | FVGen · ablation study | FID: 21.48 -> 21.05 | Improved |
| add DMD distribution matching after GAN (ablation) | FVGen · ablation study | PSNR: 16.23 -> 16.85 | Improved |
| replace reverse KL with soften reverse KL (ablation) | FVGen · ablation study | PSNR: 16.85 -> 17.50 | Improved |
| replace reverse KL with soften reverse KL (ablation) | FVGen · ablation study | FID: 21.05 -> 20.54 | Improved |
| replace reverse KL with soften reverse KL (ablation) | FVGen · ablation study | SSIM: 0.385 -> 0.382 | Hurt |
| distill ViewCrafter into 4 steps vs 66.3s teacher | FVGen · MipNeRF360 | sampling time (s): 66.3 -> 5.1 | Improved |
| FVGen vs CausVid distillation of ViewCrafter | FVGen · MipNeRF360 | PSNR: 15.77 -> 16.28 | Improved |
| FVGen vs DMD2 distillation of ViewCrafter | FVGen · MipNeRF360 | FID: 35.59 -> 17.44 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- ViewCrafter — ViewCrafter: Taming Video Diffusion Models for High-fidelity Novel View Synthesis (TPAMI 2025) · better on 7 of 10 shared results
- DepthSplat — DepthSplat: Connecting Gaussian Splatting and Depth (CVPR 2025) · better on 2 of 2 shared results
- MVSplat360 — MVSplat360: Feed-Forward 360 Scene Synthesis from Sparse Views (NeurIPS 2024) · better on 2 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
- SF-V: Single Forward Video Generation Model (NeurIPS 2024 · similarity 0.90)
- NVS-Solver: Video Diffusion Model as Zero-Shot Novel View Synthesizer (ICLR 2025 · similarity 0.89)
- CAD: Photorealistic 3D Generation via Adversarial Distillation (CVPR 2024 · similarity 0.89)
- OSV: One Step is Enough for High-Quality Image to Video Generation (CVPR 2025 · similarity 0.89)
- ViVid-1-to-3: Novel View Synthesis with Video Diffusion Models (CVPR 2024 · similarity 0.89)
- GenFusion: Closing the Loop between Reconstruction and Generation via Videos (CVPR 2025 · similarity 0.89)
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.
