Neural Adaptive SCEne Tracing
This addresses the bottleneck of slow reconstruction training for neural rendering applications, particularly for complex scenes like outdoor UAV captures, though it appears incremental as it builds on existing neural representation paradigms.
The paper tackles the problem of slow training times in neural scene reconstruction by introducing Neural Adaptive Scene Tracing (NAScenT), a hybrid explicit-implicit neural representation method that reduces training time while maintaining or improving quality compared to existing approaches.
Neural rendering with implicit neural networks has recently emerged as an attractive proposition for scene reconstruction, achieving excellent quality albeit at high computational cost. While the most recent generation of such methods has made progress on the rendering (inference) times, very little progress has been made on improving the reconstruction (training) times. In this work, we present Neural Adaptive Scene Tracing (NAScenT), the first neural rendering method based on directly training a hybrid explicit-implicit neural representation. NAScenT uses a hierarchical octree representation with one neural network per leaf node and combines this representation with a two-stage sampling process that concentrates ray samples where they matter most near object surfaces. As a result, NAScenT is capable of reconstructing challenging scenes including both large, sparsely populated volumes like UAV captured outdoor environments, as well as small scenes with high geometric complexity. NAScenT outperforms existing neural rendering approaches in terms of both quality and training time.