ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 2311.07044
29
4

L0L_0L0​-Sampler: An L0L_{0}L0​ Model Guided Volume Sampling for NeRF

13 November 2023
Liangchen Li
Juyong Zhang
ArXivPDFHTML
Abstract

Since being proposed, Neural Radiance Fields (NeRF) have achieved great success in related tasks, mainly adopting the hierarchical volume sampling (HVS) strategy for volume rendering. However, the HVS of NeRF approximates distributions using piecewise constant functions, which provides a relatively rough estimation. Based on the observation that a well-trained weight function w(t)w(t)w(t) and the L0L_0L0​ distance between points and the surface have very high similarity, we propose L0L_0L0​-Sampler by incorporating the L0L_0L0​ model into w(t)w(t)w(t) to guide the sampling process. Specifically, we propose to use piecewise exponential functions rather than piecewise constant functions for interpolation, which can not only approximate quasi-L0L_0L0​ weight distributions along rays quite well but also can be easily implemented with few lines of code without additional computational burden. Stable performance improvements can be achieved by applying L0L_0L0​-Sampler to NeRF and its related tasks like 3D reconstruction. Code is available at https://ustc3dv.github.io/L0-Sampler/ .

View on arXiv
Comments on this paper