通过智能选择像素监督，大幅降低NeRF训练的显存与时间开销。
Expansive Supervision for Neural Radiance Field
- 基于图像内容选择关键像素进行监督，避免全图渲染。
- 在保持视觉质量前提下，显存减少52%，训练时间节省16%。
- 适合追求高效训练的3D重建与视觉生成研究者。
Neural Radiance Field (NeRF) 在生成沉浸式媒体表示方面取得了显著成功，但其训练过程中密集前向传播和体素渲染带来的计算开销仍制约实际应用。本文提出 Expansive Supervision，从部分光线选择监督的角度降低训练时的时间与内存消耗。我们观察到训练误差呈现长尾分布，与图像内容相关。基于此，方法在每轮迭代中仅选择一小部分关键像素进行渲染，并将这些像素的误差值扩展至整幅图像以估计全局误差。相比传统监督方式，该方法有效规避了冗余渲染过程，显著降低时间和内存开销。实验表明，在现有最先进的加速框架中集成 Expansive Supervision，可实现52%的内存节省和16%的时间节省，同时保持相近的视觉质量。
原文摘要 · Abstract (English)
Neural Radiance Field (NeRF) has achieved remarkable success in creating immersive media representations through its exceptional reconstruction capabilities. However, the computational demands of dense forward passes and volume rendering during training continue to challenge its real-world applications. In this paper, we introduce Expansive Supervision to reduce time and memory costs during NeRF training from the perspective of partial ray selection for supervision. Specifically, we observe that training errors exhibit a long-tail distribution correlated with image content. Based on this observation, our method selectively renders a small but crucial subset of pixels and expands their values to estimate errors across the entire area for each iteration. Compared to conventional supervision, our approach effectively bypasses redundant rendering processes, resulting in substantial reductions in both time and memory consumption. Experimental results demonstrate that integrating Expansive Supervision within existing state-of-the-art acceleration frameworks achieves 52% memory savings and 16% time savings while maintaining comparable visual quality.
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。