EfficientNeRF: Efficient Neural Radiance Fields
This work addresses efficiency bottlenecks in NeRF for 3D scene representation, making it more practical for real-world applications, though it is incremental as it builds on existing NeRF methods.
The paper tackles the slow training and rendering times of Neural Radiance Fields (NeRF) by proposing EfficientNeRF, which reduces training time by over 88% and achieves rendering speeds of over 200 FPS while maintaining competitive accuracy.
Neural Radiance Fields (NeRF) has been wildly applied to various tasks for its high-quality representation of 3D scenes. It takes long per-scene training time and per-image testing time. In this paper, we present EfficientNeRF as an efficient NeRF-based method to represent 3D scene and synthesize novel-view images. Although several ways exist to accelerate the training or testing process, it is still difficult to much reduce time for both phases simultaneously. We analyze the density and weight distribution of the sampled points then propose valid and pivotal sampling at the coarse and fine stage, respectively, to significantly improve sampling efficiency. In addition, we design a novel data structure to cache the whole scene during testing to accelerate the rendering speed. Overall, our method can reduce over 88\% of training time, reach rendering speed of over 200 FPS, while still achieving competitive accuracy. Experiments prove that our method promotes the practicality of NeRF in the real world and enables many applications.