Sharpening Your Density Fields: Spiking Neuron Aided Fast Geometry Learning
This work addresses a bottleneck in NeRF geometry extraction for real-world applications, though it is incremental as it builds upon existing threshold-based methods.
The paper tackled the problem of laborious and scenario-specific threshold tuning in geometry extraction from Neural Radiance Fields (NeRF) by introducing a spiking neuron mechanism with a round-robin strategy, resulting in a more robust and efficient solution that achieves sharper and more precise density distribution with minimal computational overhead.
Neural Radiance Fields (NeRF) have achieved remarkable progress in neural rendering. Extracting geometry from NeRF typically relies on the Marching Cubes algorithm, which uses a hand-crafted threshold to define the level set. However, this threshold-based approach requires laborious and scenario-specific tuning, limiting its practicality for real-world applications. In this work, we seek to enhance the efficiency of this method during the training time. To this end, we introduce a spiking neuron mechanism that dynamically adjusts the threshold, eliminating the need for manual selection. Despite its promise, directly training with the spiking neuron often results in model collapse and noisy outputs. To overcome these challenges, we propose a round-robin strategy that stabilizes the training process and enables the geometry network to achieve a sharper and more precise density distribution with minimal computational overhead. We validate our approach through extensive experiments on both synthetic and real-world datasets. The results show that our method significantly improves the performance of threshold-based techniques, offering a more robust and efficient solution for NeRF geometry extraction.