{"slug": "toward-high-fidelity-3d-point-cloud-learning-for-brain-folding-morphology-using", "title": "Toward High-Fidelity 3D Point-Cloud Learning for Brain Folding Morphology Prediction Using Trans-Unet", "summary": "Researchers propose Trans-Unet, a novel framework that transforms 3D point-cloud data into a 2D grid domain and uses a U-shaped hybrid model integrating CNNs and self-attention, to predict brain folding morphology from high-resolution 3D point-cloud data with 40,401 surface points and 2,382 fiber points. The method achieves high-fidelity predictions of brain patch growth, outperforming existing approaches in accuracy and fidelity.", "body_md": "arXiv:2607.21840v1 Announce Type: new\nAbstract: Learning high-fidelity point-cloud features in the 3D space poses significant challenges, including permutation invariance, lack of local context, difficulty in fine-grained surface reconstruction, and high computational cost. In this article, we propose Trans-Unet, a novel framework that addresses these issues by first tansforming 3D point-cloud data into a 2D grid domain and then employing a U-shaped hybrid model that integrates Convolutional Neural Networks, and self-attention mechanisms. The proposed Trans-Unet effectively learns and reconstructs precise features from high-resolution 3D point-cloud data (with 40,401 points in surface and 2,382 points in fiber) derived from a predefined finite element brain patch growth model, enabling accurate prediction of brain folding patterns. By combining multiple techniques, Trans-Unet leverages the complementary strengths: the 3D-to-2D transformation preserves fine-grained structural information while significantly reducing computational cost and the curse of dimensionality; convolutional blocks capture hierarchical, low-level local representations; and the self-attention mechanism models global, high-level semantics and long-range dependencies. The dataset consists of 3D point-clouds containing both brain surface patches and fiber information generated by a large-scale finite element model. Trans-Unet is applied to predict brain surface folding from the initial state (state 0 or states 0-2) to the final state (state 3). Experimental results demonstrate that Trans-Unet achieves high-resolution predictions of brain patch growth, surpassing existing methods in both fidelity and accuracy.", "url": "https://wpnews.pro/news/toward-high-fidelity-3d-point-cloud-learning-for-brain-folding-morphology-using", "canonical_source": "https://arxiv.org/abs/2607.21840", "published_at": "2026-07-27 04:00:00+00:00", "updated_at": "2026-07-27 04:27:26.037195+00:00", "lang": "en", "topics": ["machine-learning", "computer-vision"], "entities": ["Trans-Unet"], "alternates": {"html": "https://wpnews.pro/news/toward-high-fidelity-3d-point-cloud-learning-for-brain-folding-morphology-using", "markdown": "https://wpnews.pro/news/toward-high-fidelity-3d-point-cloud-learning-for-brain-folding-morphology-using.md", "text": "https://wpnews.pro/news/toward-high-fidelity-3d-point-cloud-learning-for-brain-folding-morphology-using.txt", "jsonld": "https://wpnews.pro/news/toward-high-fidelity-3d-point-cloud-learning-for-brain-folding-morphology-using.jsonld"}}