# MCBlock: Boosting Neural Radiance Field Training Speed by MCTS-based Dynamic-Resolution Ray Sampling

arXiv: https://arxiv.org/abs/2504.09878 · HTML: https://mlsanity.com/paper/2504.09878/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 25 of its 44 references that are in the map.
- First posted on arXiv 14 Apr 2025.

## Builds on (most cited first, 25 of 25)

- [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023)
- [Instant Neural Graphics Primitives with a Multiresolution Hash Encoding](https://mlsanity.com/paper/2201.05989/) (2022)
- [Mip-NeRF: A Multiscale Representation for Anti-Aliasing Neural Radiance Fields](https://mlsanity.com/paper/2103.13415/) (2021)
- [Plenoxels: Radiance Fields without Neural Networks](https://mlsanity.com/paper/2112.05131/) (2021)
- [pixelNeRF: Neural Radiance Fields from One or Few Images](https://mlsanity.com/paper/2012.02190/) (2020)
- [Direct Voxel Grid Optimization: Super-fast Convergence for Radiance Fields Reconstruction](https://mlsanity.com/paper/2111.11215/) (2021)
- [Neural Sparse Voxel Fields](https://mlsanity.com/paper/2007.11571/) (2020)
- [GRAF: Generative Radiance Fields for 3D-Aware Image Synthesis](https://mlsanity.com/paper/2007.02442/) (2020)
- [Nerfstudio: A Modular Framework for Neural Radiance Field Development](https://mlsanity.com/paper/2302.04264/) (2023)
- [NeRV: Neural Reflectance and Visibility Fields for Relighting and View Synthesis](https://mlsanity.com/paper/2012.03927/) (2020)
- [Instruct-NeRF2NeRF: Editing 3D Scenes with Instructions](https://mlsanity.com/paper/2303.12789/) (2023)
- [Introduction to Multi-Armed Bandits](https://mlsanity.com/paper/1904.07272/) (2019)
- [NeRD: Neural Reflectance Decomposition from Image Collections](https://mlsanity.com/paper/2012.03918/) (2020)
- [Editable Free-viewpoint Video Using a Layered Neural Representation](https://mlsanity.com/paper/2104.14786/) (2021)
- [SINE: Semantic-driven Image-based NeRF Editing with Prior-guided Editing Field](https://mlsanity.com/paper/2303.13277/) (2023)
- [Real-time Neural Radiance Caching for Path Tracing](https://mlsanity.com/paper/2106.12372/) (2021)
- [Local Implicit Ray Function for Generalizable Radiance Field Representation](https://mlsanity.com/paper/2304.12746/) (2023)
- [Accelerating Neural Field Training via Soft Mining](https://mlsanity.com/paper/2312.00075/) (2023)
- [Watch the Unobserved: A Simple Approach to Parallelizing Monte Carlo Tree Search](https://mlsanity.com/paper/1810.11755/) (2018)
- [Know Your Neighbors: Improving Single-View Reconstruction via Spatial Vision-Language Reasoning](https://mlsanity.com/paper/2404.03658/) (2024)
- [Efficient Ray Sampling for Radiance Fields Reconstruction](https://mlsanity.com/paper/2308.15547/) (2023)
- [Improving Neural Radiance Field using Near-Surface Sampling with Point Cloud Generation](https://mlsanity.com/paper/2310.04152/) (2023)
- [Efficient NeRF Optimization -- Not All Samples Remain Equally Hard](https://mlsanity.com/paper/2408.03193/) (2024)
- [GL-NeRF: Gauss-Laguerre Quadrature Enables Training-Free NeRF Acceleration](https://mlsanity.com/paper/2410.19831/) (2024)
- [Plug-and-Play Acceleration of Occupancy Grid-based NeRF Rendering using VDB Grid and Hierarchical Ray Traversal](https://mlsanity.com/paper/2404.10272/) (2024)

## Abstract (arXiv)

Neural Radiance Field (NeRF) is widely known for high-fidelity novel view synthesis. However, even the state-of-the-art NeRF model, Gaussian Splatting, requires minutes for training, far from the real-time performance required by multimedia scenarios like telemedicine. One of the obstacles is its inefficient sampling, which is only partially addressed by existing works. Existing point-sampling algorithms uniformly sample simple-texture regions (easy to fit) and complex-texture regions (hard to fit), while existing ray-sampling algorithms sample these regions all in the finest granularity (i.e. the pixel level), both wasting GPU training resources. Actually, regions with different texture intensities require different sampling granularities. To this end, we propose a novel dynamic-resolution ray-sampling algorithm, MCBlock, which employs Monte Carlo Tree Search (MCTS) to partition each training image into pixel blocks with different sizes for active block-wise training. Specifically, the trees are initialized according to the texture of training images to boost the initialization speed, and an expansion/pruning module dynamically optimizes the block partition. MCBlock is implemented in Nerfstudio, an open-source toolset, and achieves a training acceleration of up to 2.33x, surpassing other ray-sampling algorithms. We believe MCBlock can apply to any cone-tracing NeRF model and contribute to the multimedia community.