Research map / 3D vision / Gaussian splatting
TokenSplat: Token-aligned 3D Gaussian Splatting for Feed-forward Pose-free Reconstruction
Yihui Li, Chengxin Lv, Zichen Tang, Hongyu Yang, Di 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 |
|---|---|---|---|
| use token-aligned Gaussian prediction instead of a pixel-aligned head (ablation) | TokenSplat · RE10K (8 views) | PSNR (dB): 25.33 -> 26.15 | Improved |
| use token-aligned Gaussian prediction instead of a pixel-aligned head (ablation) | TokenSplat · RE10K (8 views) | RPE-r: 0.496 -> 0.458 | Improved |
| use token-based fusion instead of AnySplat-style Gaussian fusion (ablation) | TokenSplat · RE10K (8 views) | PSNR (dB): 25.77 -> 26.15 | Improved |
| add the ADF-Decoder instead of a standard ViT decoder (ablation) | TokenSplat · RE10K (8 views) | PSNR (dB): 25.88 -> 26.15 | Improved |
| add the ADF-Decoder instead of a standard ViT decoder (ablation) | TokenSplat · RE10K (8 views) | RPE-r: 0.504 -> 0.458 | Improved |
| add the ADF-Decoder instead of a standard ViT decoder (ablation) | TokenSplat · RE10K (8 views) | ATE: 0.023 -> 0.012 | Improved |
| add the known intrinsic camera embedding (ablation) | TokenSplat · RE10K (8 views) | PSNR (dB): 25.54 -> 26.15 | Improved |
| add the known intrinsic camera embedding (ablation) | TokenSplat · RE10K (8 views) | RPE-r: 0.471 -> 0.458 | No clear effect |
| use TokenSplat instead of pose-required FreeSplat | FreeSplat · RE10K (8 views) | PSNR (dB): 25.20 -> 26.15 | Improved |
| increase test views to 28 using a model trained with 10 views | TokenSplat · multi-view Table 2 | PSNR (dB): 26.83 -> 26.87 | No clear effect |
| increase test views from 10 to 28 | FreeSplat · multi-view Table 2 | PSNR (dB): 25.38 -> 24.30 | Hurt |
| increase test views from 10 to 28 | NoPoSplat · multi-view Table 2 | PSNR (dB): 24.21 -> 23.21 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- FreeSplat — FreeSplat: Generalizable 3D Gaussian Splatting Towards Free View Synthesis of Indoor Scenes (NeurIPS 2024) · better on 4 of 4 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
- No Pose, No Problem: Surprisingly Simple 3D Gaussian Splats from Sparse Unposed Images (ICLR 2025 · similarity 0.90)
- SelfSplat: Pose-Free and 3D Prior-Free Generalizable 3D Gaussian Splatting (CVPR 2025 · similarity 0.88)
- PF3plat: Pose-Free Feed-Forward 3D Gaussian Splatting for Novel View Synthesis (ICML 2025 · similarity 0.88)
- TokenHMR: Advancing Human Mesh Recovery with a Tokenized Pose Representation (CVPR 2024 · similarity 0.86)
- OnlineSplatter: Pose-Free Online 3D Reconstruction for Free-Moving Objects (NeurIPS 2025 · similarity 0.86)
- FreeSplat: Generalizable 3D Gaussian Splatting Towards Free View Synthesis of Indoor Scenes (NeurIPS 2024 · 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.
