Research map / 3D vision / Face and shape
MinD-3D++: Advancing fMRI-Based 3D Reconstruction With High-Quality Textured Mesh Generation and a Comprehensive Dataset
Jianxiong Gao, Yanwei Fu, Yuqian Fu, Yun Wang, Xuelin Qian, Jianfeng Feng
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 contrastive loss (ablation) | MinD-3D++ · fMRI-Shape (Core set) | 10-way accuracy: 0.463 -> 0.616 | Improved |
| add contrastive loss (ablation) | MinD-3D++ · fMRI-Shape (Core set) | LPIPS: 0.296 -> 0.234 | Improved |
| add image branch to contrastive loss (ablation) | MinD-3D++ · fMRI-Shape (Core set) | 10-way accuracy: 0.549 -> 0.616 | Improved |
| add text branch to contrastive loss (ablation) | MinD-3D++ · fMRI-Shape (Core set) | 10-way accuracy: 0.563 -> 0.616 | Improved |
| add image branch to contrastive loss (ablation) | MinD-3D++ · fMRI-Shape (Core set) | PSNR: 15.79 -> 16.44 | Improved |
| add text branch to contrastive loss (ablation) | MinD-3D++ · fMRI-Shape (Core set) | SSIM: 0.734 -> 0.763 | Improved |
| apply contrastive loss at intermediate layer instead of final 12th layer (ablation) | MinD-3D++ · fMRI-Shape (Core set) | 10-way accuracy: 0.604 -> 0.616 | Improved |
| apply contrastive loss at intermediate layer instead of final 12th layer (ablation) | MinD-3D++ · fMRI-Shape (Core set) | SSIM: 0.759 -> 0.763 | Improved |
| replace MinD-3D with MinD-3D++ (texture-aware) | MinD-3D · fMRI-Shape (Core set) | LPIPS: 0.306 -> 0.234 | Improved |
| replace MinD-3D with MinD-3D++ on larger dataset | MinD-3D · fMRI-Objaverse | 10-way accuracy: 0.427 -> 0.618 | Improved |
| replace MinD-3D with MinD-3D++ in across-person and across-class test | MinD-3D · fMRI-Shape APAC | FPD: 5.689 -> 5.347 | Improved |
| replace MinD-3D with MinD-3D++ in across-person test | MinD-3D · fMRI-Shape AP | FPD: 3.838 -> 3.582 | Improved |
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.
- MinD-3D — MinD-3D: Reconstruct High-quality 3D objects in Human Brain (ECCV 2024) · better on 10 of 10 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
- MindEye2: Shared-Subject Models Enable fMRI-To-Image With 1 Hour of Data (ICML 2024 · similarity 0.90)
- NSD-Imagery: A Benchmark Dataset for Extending fMRI Vision Decoding Methods to Mental Imagery (CVPR 2025 · similarity 0.89)
- Neuro-Vision to Language: Enhancing Brain Recording-based Visual Reconstruction and Language Interaction (NeurIPS 2024 · similarity 0.88)
- Neuro-3D: Towards 3D Visual Decoding from EEG Signals (CVPR 2025 · similarity 0.87)
- FlatClip: Reusing Image Foundation Models for fMRI Representation Learning via Cortical Flatmaps (NeurIPS 2026 · similarity 0.87)
- MindBridge: A Cross-Subject Brain Decoding Framework (CVPR 2024 · 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.
