arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2603.20128cs.CV
收藏

可推广的NGP-SR:通过神经图形原语实现可推广的神经辐射场超分辨率

Generalizable NGP-SR: Generalizable Neural Radiance Fields Super-Resolution via Neural Graph Primitives

Clemson University(克莱姆森大学)

Wanqi Yuan, Omkar Sharad Mayekar, Connor Pennington, Nianyi Li

AI总结:

本文提出一种3D感知的超分辨率框架,直接从低分辨率图像重建高分辨率辐射场,无需外部参考或后处理,提升重建质量和运行效率。

摘要:

神经辐射场(NeRF)能够实现逼真的新视角合成,但在需要高分辨率(HR)渲染时成本较高,因为HR输出需要密集采样和更高容量的模型。此外,简单地在2D上每个视角进行超分辨率处理会破坏多视角一致性。我们提出了可推广的NGP-SR,一种3D感知的超分辨率框架,直接从低分辨率(LR)姿态图像重建高分辨率辐射场。基于神经图形原语(NGP),NGP-SR将辐射预测条件设置为3D坐标和学习的局部纹理标记,使在辐射场中恢复高频细节,并在不使用外部HR参考或后处理2D上采样的情况下产生视图一致的HR新视角。重要的是,我们的模型具有可推广性:一旦训练完成,它可以应用于未见过的场景,并从新视角渲染,而无需每场景优化。在多个数据集上的实验表明,NGP-SR在重建质量和运行效率上都优于先前基于NeRF的超分辨率方法,提供了一种可扩展的高分辨率新视角合成的实用解决方案。

英文摘要:

Neural Radiance Fields (NeRF) achieve photorealistic novel view synthesis but become costly when high-resolution (HR) rendering is required, as HR outputs demand dense sampling and higher-capacity models. Moreover, naively super-resolving per-view renderings in 2D often breaks multi-view consistency. We propose Generalizable NGP-SR, a 3D-aware super-resolution framework that reconstructs an HR radiance field directly from low-resolution (LR) posed images. Built on Neural Graphics Primitives (NGP), NGP-SR conditions radiance prediction on 3D coordinates and learned local texture tokens, enabling recovery of high-frequency details within the radiance field and producing view-consistent HR novel views without external HR references or post-hoc 2D upsampling. Importantly, our model is generalizable: once trained, it can be applied to unseen scenes and rendered from novel viewpoints without per-scene optimization. Experiments on multiple datasets show that NGP-SR consistently improves both reconstruction quality and runtime efficiency over prior NeRF-based super-resolution methods, offering a practical solution for scalable high-resolution novel view synthesis.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑