Research map / 3D vision / Gaussian splatting
More Natural, More Real: Object-aware Gaussian Splatting for 3D Visual Decoding from Human Brain
Haodong Jing, Dongyao Jiang, Jixin Wang, Junhao Jia, Yanshu Li, Yongqiang Ma, Nanning Zheng
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 Multi-Attribute Controller (ablation) | BrainGS · fMRI-Shape | CD: 2.780 -> 1.517 | Improved |
| add Multi-View Stabilizer (ablation) | BrainGS · fMRI-Shape | FPD: 3.795 -> 2.936 | Improved |
| add Object-aware Enhancement (ablation) | BrainGS · fMRI-Shape | FPD: 3.015 -> 2.936 | Improved |
| add View-Guided Adjustment (ablation) | BrainGS · fMRI-Shape | FPD: 3.171 -> 2.936 | Improved |
| use FTS-Net instead of LEA fMRI encoder (ablation) | BrainGS · fMRI-Shape | FPD: 3.011 -> 2.936 | Improved |
| add color loss Lc (ablation) | BrainGS · EEG-3D color classification | top-1 accuracy: 24.52 -> 38.18 | Improved |
| add view consistency loss Lview (ablation) | BrainGS · EEG-3D object classification | top-1 accuracy: 5.05 -> 6.12 | Improved |
| add alignment loss Lalign (ablation) | BrainGS · EEG-3D object classification | top-1 accuracy: 6.12 -> 6.48 | Improved |
| deform all Gaussian attributes instead of only SH and opacity (ablation) | BrainGS · fMRI-Shape | FPD: 3.018 -> 2.936 | Improved |
| deform all Gaussian attributes instead of only position, rotation, scale (ablation) | BrainGS · fMRI-Shape | FPD: 3.528 -> 2.936 | Improved |
| use canonical Gaussian initialization instead of random init (ablation) | BrainGS · fMRI-Shape | FPD: 4.177 -> 2.936 | Improved |
| use all brain regions instead of a single lobe (ablation) | BrainGS · fMRI-Shape | FPD: 3.226 -> 2.936 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- MinD-3D++ — MinD-3D: Reconstruct High-quality 3D objects in Human Brain (ECCV 2024) · better on 5 of 5 shared results
- Neuro-3D — Neuro-3D: Towards 3D Visual Decoding from EEG Signals (CVPR 2025) · 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
- Neuro-3D: Towards 3D Visual Decoding from EEG Signals (CVPR 2025 · similarity 0.91)
- Neural Signed Distance Function Inference through Splatting 3D Gaussians Pulled on Zero-Level Set (NeurIPS 2024 · similarity 0.88)
- Binocular-Guided 3D Gaussian Splatting with View Consistency for Sparse View Synthesis (NeurIPS 2024 · similarity 0.88)
- 3DGS-Enhancer: Enhancing Unbounded 3D Gaussian Splatting with View-consistent 2D Diffusion Priors (NeurIPS 2024 · similarity 0.88)
- 3D Student Splatting and Scooping (CVPR 2025 · similarity 0.87)
- Free360: Layered Gaussian Splatting for Unbounded 360-Degree View Synthesis from Extremely Sparse and Unposed Views (CVPR 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.
