Research map / 3D vision / Gaussian splatting
Forge4D: Feed-Forward 4D Human Reconstruction and Interpolation from Uncalibrated Sparse-View Videos
Yingdong Hu, YISHENG HE, Jinnan Chen, Weihao Yuan, Kejie Qiu, Zehong Lin, Siyu Zhu, Zilong Dong, Steven Hoi, Jun Zhang
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 metric gauge aligning (ablation) | Forge4D · DNA-Rendering static novel view | PSNR: 13.2884 -> 29.8167 | Improved |
| add state token (ablation) | Forge4D · DNA-Rendering novel-time + novel-view | PSNR: 28.5555 -> 29.0378 | Improved |
| add retargeting loss (ablation) | Forge4D · DNA-Rendering novel-time + novel-view | PSNR: 28.4124 -> 29.0378 | Improved |
| add occlusion-aware optical flow loss (ablation) | Forge4D · DNA-Rendering novel-time + novel-view | PSNR: 28.8676 -> 29.0378 | Improved |
| add Gaussian Fusion instead of plain averaging (ablation) | Forge4D · DNA-Rendering novel-time + novel-view | PSNR: 27.3459 -> 29.0378 | Improved |
| add Gaussian Fusion (ablation) | Forge4D · DNA-Rendering novel-time + novel-view | TL-STD: 0.9446 -> 0.5379 | Improved |
| use 2 input cameras instead of 4 (ablation) | Forge4D · DNA-Rendering novel-time + novel-view | PSNR: 29.0378 -> 27.1316 | Hurt |
| use 5 input cameras instead of 4 (ablation) | Forge4D · DNA-Rendering novel-time + novel-view | PSNR: 29.0378 -> 28.1337 | Hurt |
| increase sampling rate from 2 to 4 (ablation) | Forge4D · DNA-Rendering novel-time + novel-view | PSNR: 29.0378 -> 27.9129 | Hurt |
| increase sampling rate from 2 to 8 (ablation) | Forge4D · DNA-Rendering novel-time + novel-view | PSNR: 29.0378 -> 25.9673 | Hurt |
| increase video length from 8 to 16 timestamps (ablation) | Forge4D · DNA-Rendering novel-time + novel-view | PSNR: 29.0378 -> 29.3932 | No clear effect |
| increase video length from 8 to 32 timestamps (ablation) | Forge4D · DNA-Rendering novel-time + novel-view | PSNR: 29.0378 -> 29.3615 | No clear effect |
Later papers that compared with this method
1 later accepted paper reports Forge4D as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- FA-LAM: Focus-Aware Large Avatar Model for One-Shot 4D Animatable Gaussian Head (NeurIPS 2026 · better on 1 of 1 shared result)
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.
- GPS-Gaussian — GPS-Gaussian: Generalizable Pixel-wise 3D Gaussian Splatting for Real-time Human Novel View Synthesis (CVPR 2024) · better on 3 of 3 shared results
- L4GM — L4GM: Large 4D Gaussian Reconstruction Model (NeurIPS 2024) · better on 2 of 2 shared results
- MoGe-2 — MoGe-2: Accurate Monocular Geometry with Metric Scale and Sharp Details (NeurIPS 2025) · better on 1 of 1 shared result
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
- ODHSR: Online Dense 3D Reconstruction of Humans and Scenes from Monocular Videos (CVPR 2025 · similarity 0.89)
- Guess The Unseen: Dynamic 3D Scene Reconstruction from Partial 2D Glimpses (CVPR 2024 · similarity 0.88)
- 4D-LRM: Large Space-Time Reconstruction Model From and To Any View at Any Time (NeurIPS 2025 · similarity 0.88)
- Instant4D: 4D Gaussian Splatting in Minutes (NeurIPS 2025 · similarity 0.87)
- 4D3R: Motion-Aware Neural Reconstruction and Rendering of Dynamic Scenes from Monocular Videos (NeurIPS 2025 · similarity 0.87)
- UniCon3R: Unified Contact-aware 4D Human-Scene Reconstruction from Monocular Video (NeurIPS 2026 · 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.
