Research map / 3D vision / 3D generation
Brain3D: Generating 3D Objects from fMRI
Yuankun Yang, Li Zhang, Ziyang Xie, Zhiyuan Yuan, Jianfeng Feng, Xiatian Zhu, Yu–Gang Jiang
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 UMAP projection of high-level fMRI embeddings (ablation) | Brain3D · NSD | 3D Inception/CLIP/EfficientNet/Swav: higher 3D Inception and 3D CLIP, lower 3D EfficientNet and 3D Swav | Improved |
| add semantic stage (DMTet mesh refinement) after perceptual NeRF stage (ablation) | Brain3D · NSD | 3D Inception/CLIP/EfficientNet/Swav | Improved |
| use both low-level and high-level guidance instead of either alone (ablation) | Brain3D perceptual stage · NSD | 3D metrics | Improved |
| combine left and right hemisphere fMRI instead of one hemisphere (ablation) | Brain3D · NSD | 3D metrics | Improved |
| set UMAP N_neighbor to 50 instead of 1 (ablation) | Brain3D UMAP projection · NSD | qualitative 3D geometry and texture: N_neighbor=1 vs 50 | Improved |
| increase UMAP N_neighbor from 50 to 200 or 1000 (ablation) | Brain3D UMAP projection · NSD | qualitative texture: 50 -> 200 and 1000 | No clear effect |
| use visual cortex fMRI instead of MTL fMRI (ablation) | Brain3D · NSD | 3D Inception p-value: p=1e-23 | Improved |
| use fewer diagnostic images (10 instead of 100) (ablation) | Brain3D diagnosis V1 and MTL · NSD simulated disorders | diagnostic accuracy: over 95% accuracy with 10 images | No clear effect |
| use Brain3D 3D outputs instead of 2D MindEye for diagnosis | clinical diagnosis · NSD simulated disorders | diagnosis performance | Improved |
| add synthetic Stable Diffusion pseudo image prior to text-imagination fMRI (ablation) | Brain3D · TMNRED | qualitative detail | Improved |
Closest papers by meaning
- Neuro-3D: Towards 3D Visual Decoding from EEG Signals (CVPR 2025 · similarity 0.90)
- Neuro-Vision to Language: Enhancing Brain Recording-based Visual Reconstruction and Language Interaction (NeurIPS 2024 · similarity 0.89)
- MindEye2: Shared-Subject Models Enable fMRI-To-Image With 1 Hour of Data (ICML 2024 · similarity 0.88)
- Brain decoding: toward real-time reconstruction of visual perception (ICLR 2024 · similarity 0.88)
- A Convolutional Neural Network Interpretable Framework for Human Ventral Visual Pathway Representation (AAAI 2024 · similarity 0.88)
- NSD-Imagery: A Benchmark Dataset for Extending fMRI Vision Decoding Methods to Mental Imagery (CVPR 2025 · similarity 0.88)
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.
