arXiv cs.AI· Chuanzhi Xu, Boyu Wei, Chengkun Yue, Xuanhua Yin, Yiyang Li, Haodong Chen, Qiang Qu, Weidong Cai·· 23 小时前AI 评分24
ReTouchGS:面向 3D Gaussian Splatting 的超轻量免渲染美学色彩增强与迁移
ReTouchGS: Ultralight Render-free Aesthetic Color Enhancement and Transfer for 3D Gaussian Splatting
AI 导读
ReTouchGS 是首个面向预训练 3D Gaussian Splatting 模型的免参考美学色彩增强框架,可从模型及其相机参数预测场景级 LUT。它同时支持参考引导的色彩迁移,仅用 44.7K 参数的混合感知色彩细化网络在无需渲染的情况下完成色彩映射,数秒内输出标准 3DGS 资产并保持几何不变。团队还构建了 ReTouch3D benchmark 用于跨视角评估。
正文 · 原文
Abstract:3D Gaussian Splatting (3DGS) enables photorealistic scene reconstruction. Existing 3DGS color transfer and interactive editing methods are guided by reference images and user-specified color adjustments. However, in many applications, users simply want aesthetically guided color enhancement that automatically makes the scene more visually appealing while maintaining consistent colors and tones across viewpoints. We present ReTouchGS, to our knowledge the first framework for reference-free aesthetic color enhancement of pretrained 3DGS models. It predicts a scene-level color lookup table (LUT) from the model and its associated camera parameters. ReTouchGS also supports reference-guided color transfer by fitting a scene-level LUT using one edited view of the same scene and its corresponding original render. A shared blend-aware color refinement network with only 44.7K parameters refines an analytic color update to realize either color mapping without rendering. It preserves geometry and delivers standard 3DGS assets in seconds. We also design the ReTouch3D benchmark to support the evaluation of aesthetic color enhancement and reference-guided color transfer across viewpoints. Experiments demonstrate high color accuracy and strong generalization to unseen scenes in both aesthetic color enhancement and reference-guided color transfer.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Graphics (cs.GR) |
| Cite as: | arXiv:2611.23054 [cs.CV] |
| (or arXiv:2611.23054v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2611.23054 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Chuanzhi Xu [view email]
[v1]
Tue, 17 Nov 2026 16:47:43 UTC (4,105 KB)
来源:arXiv cs.AI · arxiv.org