{"slug": "factorsplat-appearance-controllable-gaussian-proxies-for-medical-volume", "title": "FactorSplat: Appearance-Controllable Gaussian Proxies for Medical Volume Rendering", "summary": "FactorSplat, a per-scene N-dimensional Gaussian splatting proxy from researchers behind the arXiv paper 2610.02382v1, accepts region-specific intensity-to-RGBA transfer function curves at inference, letting one checkpoint per scan support unseen appearance edits without retraining. Across seven CT and MR scans and four splits (validation, interpolation, unseen composition, out-of-distribution edits), FactorSplat improved mean PSNR over region-aware VEG by 1.10 to 1.52 dB, and its cached fast renderer averages 524 FPS at 1600² with 1.17 ms transfer-function switches.", "body_md": "arXiv:2610.02382v1 Announce Type: new \nAbstract: Transfer functions (TFs) control color and visibility in medical volume rendering, but image-trained Gaussian proxies typically bake one transfer function into their appearance. We present FactorSplat, a per-scene N-dimensional Gaussian splatting (N-DGS) proxy that accepts region-specific intensity-to-RGBA curves at inference. A local lookup applies the authored color and opacity change, while a shared functional encoder and low-rank per-Gaussian factors learn the residual appearance response. Geometry and directional appearance remain shared across presets, with visibility control and TF-aware pruning preserving the ability to hide and reveal structures. On seven CT and MR scans, FactorSplat improves mean PSNR and changed-region error over region-aware VEG across validation, interpolation, unseen composition, and out-of-distribution (OOD) edits. Across these four splits, seven-scan mean PSNR gains over VEG range from 1.10 to 1.52 dB. One checkpoint per scan supports unseen edits without retraining. At $1600^2$, the cached fast renderer averages 524 FPS with 1.17 ms TF switches. Project page: https://gaozhongpai.github.io/FactorSplat/.", "url": "https://wpnews.pro/news/factorsplat-appearance-controllable-gaussian-proxies-for-medical-volume", "canonical_source": "https://arxiv.org/abs/2610.02382", "published_at": "2026-10-05 04:00:00+00:00", "updated_at": "2026-10-05 04:13:29.695552+00:00", "lang": "en", "topics": ["computer-vision", "ai-research", "machine-learning", "artificial-intelligence"], "entities": ["FactorSplat", "Gaussian splatting", "VEG", "arXiv", "CT", "MR"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/factorsplat-appearance-controllable-gaussian-proxies-for-medical-volume", "markdown": "https://wpnews.pro/news/factorsplat-appearance-controllable-gaussian-proxies-for-medical-volume.md", "text": "https://wpnews.pro/news/factorsplat-appearance-controllable-gaussian-proxies-for-medical-volume.txt", "jsonld": "https://wpnews.pro/news/factorsplat-appearance-controllable-gaussian-proxies-for-medical-volume.jsonld"}}