让一个神经渲染模型持续学习多个3D场景，不增加参数量且保持高质量渲染。
Incremental Multi-Scene Modeling via Continual Neural Graphics Primitives
- 用生成回放技术逐步融合新场景，无需旧数据。
- 8个真实场景共用一个模型，PSNR仅下降2.2%。
- 支持同一模型内多次风格修改，适合动态场景应用。
神经辐射场（NeRF）革新了3D场景的新视角逼真渲染。尽管其作为3D资源日益流行且高效，但面临可扩展性挑战：每个场景需独立建模，多场景训练时间累积增长。将多个3D场景增量式编码到单一NeRF模型中的潜力尚未被充分探索。为此，我们提出持续神经图形基元（C-NGP），一种新颖的持续学习框架，可将多个场景逐步集成至单一神经辐射场中。采用生成回放方法，C-NGP在无需访问旧数据的情况下适应新场景。实验表明，C-NGP在不增加参数量的前提下容纳多个场景，合成与真实数据集上均生成高质量新视角图像。特别地，C-NGP成功建模了Real-LLFF数据集中的全部8个场景，相较独立建模的原始NeRF仅出现2.2%的PSNR下降。此外，该模型还支持同一网络内多次风格编辑。
原文摘要 · Abstract (English)
Neural radiance fields (NeRF) have revolutionized photorealistic rendering of novel views for 3D scenes. Despite their growing popularity and efficiency as 3D resources, NeRFs face scalability challenges due to the need for separate models per scene and the cumulative increase in training time for multiple scenes. The potential for incrementally encoding multiple 3D scenes into a single NeRF model remains largely unexplored. To address this, we introduce Continual-Neural Graphics Primitives (C-NGP), a novel continual learning framework that integrates multiple scenes incrementally into a single neural radiance field. Using a generative replay approach, C-NGP adapts to new scenes without requiring access to old data. We demonstrate that C-NGP can accommodate multiple scenes without increasing the parameter count, producing high-quality novel-view renderings on synthetic and real datasets. Notably, C-NGP models all $8$ scenes from the Real-LLFF dataset together, with only a $2.2\%$ drop in PSNR compared to vanilla NeRF, which models each scene independently. Further, C-NGP allows multiple style edits in the same network.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。