# R2-Mesh: Reinforcement Learning Powered Mesh Reconstruction via Geometry and Appearance Refinement

arXiv: https://arxiv.org/abs/2408.10135 · HTML: https://mlsanity.com/paper/2408.10135/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL with diffusion and flow models, one of 15 directions in Deep reinforcement learning methods.
- Builds on 25 of its 38 references that are in the map.
- First posted on arXiv 19 Aug 2024.

## Placed in

- Deep reinforcement learning methods / [RL with diffusion and flow models](https://mlsanity.com/deeprl/rl-with-diffusion-and-flow-models/)

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

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Instant Neural Graphics Primitives with a Multiresolution Hash Encoding](https://mlsanity.com/paper/2201.05989/) (2022)
- [Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fields](https://mlsanity.com/paper/2111.12077/) (2021)
- [NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction](https://mlsanity.com/paper/2106.10689/) (2021)
- [Plenoxels: Radiance Fields without Neural Networks](https://mlsanity.com/paper/2112.05131/) (2021)
- [TensoRF: Tensorial Radiance Fields](https://mlsanity.com/paper/2203.09517/) (2022)
- [Volume Rendering of Neural Implicit Surfaces](https://mlsanity.com/paper/2106.12052/) (2021)
- [Direct Voxel Grid Optimization: Super-fast Convergence for Radiance Fields Reconstruction](https://mlsanity.com/paper/2111.11215/) (2021)
- [K-Planes: Explicit Radiance Fields in Space, Time, and Appearance](https://mlsanity.com/paper/2301.10241/) (2023)
- [Pixel2Mesh: Generating 3D Mesh Models from Single RGB Images](https://mlsanity.com/paper/1804.01654/) (2018)
- [UNISURF: Unifying Neural Implicit Surfaces and Radiance Fields for Multi-View Reconstruction](https://mlsanity.com/paper/2104.10078/) (2021)
- [Neuralangelo: High-Fidelity Neural Surface Reconstruction](https://mlsanity.com/paper/2306.03092/) (2023)
- [Extracting Triangular 3D Models, Materials, and Lighting From Images](https://mlsanity.com/paper/2111.12503/) (2021)
- [NeuS2: Fast Learning of Neural Implicit Surfaces for Multi-view Reconstruction](https://mlsanity.com/paper/2212.05231/) (2022)
- [MobileNeRF: Exploiting the Polygon Rasterization Pipeline for Efficient Neural Field Rendering on Mobile Architectures](https://mlsanity.com/paper/2208.00277/) (2022)
- [SparseNeuS: Fast Generalizable Neural Surface Reconstruction from Sparse Views](https://mlsanity.com/paper/2206.05737/) (2022)
- [Shape, Light, and Material Decomposition from Images using Monte Carlo Rendering and Denoising](https://mlsanity.com/paper/2206.03380/) (2022)
- [Flexible Isosurface Extraction for Gradient-Based Mesh Optimization](https://mlsanity.com/paper/2308.05371/) (2023)
- [ActiveNeRF: Learning where to See with Uncertainty Estimation](https://mlsanity.com/paper/2209.08546/) (2022)
- [Delicate Textured Mesh Recovery from NeRF via Adaptive Surface Refinement](https://mlsanity.com/paper/2303.02091/) (2023)
- [NeU-NBV: Next Best View Planning Using Uncertainty Estimation in Image-Based Neural Rendering](https://mlsanity.com/paper/2303.01284/) (2023)
- [NeRFMeshing: Distilling Neural Radiance Fields into Geometrically-Accurate 3D Meshes](https://mlsanity.com/paper/2303.09431/) (2023)
- [A Deep-Learning Approach For Direct Whole-Heart Mesh Reconstruction](https://mlsanity.com/paper/2102.07899/) (2021)
- [Online Next-Best-View Planner for 3D-Exploration and Inspection With a Mobile Manipulator Robot](https://mlsanity.com/paper/2203.10113/) (2022)
- [Image-based Pose Estimation and Shape Reconstruction for Robot Manipulators and Soft, Continuum Robots via Differentiable Rendering](https://mlsanity.com/paper/2302.14039/) (2023)

## Abstract (arXiv)

Mesh reconstruction from Neural Radiance Fields (NeRF) is widely used in 3D reconstruction and has been applied across numerous domains. However, existing methods typically rely solely on the given training set images, which restricts supervision to limited observations and makes it difficult to fully constrain geometry and appearance. Moreover, the contribution of each viewpoint for training is not uniform and changes dynamically during the optimization process, which can result in suboptimal guidance for both geometric refinement and rendering quality. To address these limitations, we propose R2-Mesh, a reinforcement learning framework that combines NeRF-rendered pseudo-supervision with online viewpoint selection. Our key insight is to exploit NeRF's rendering ability to synthesize additional high-quality images, enriching training with diverse viewpoint information. To ensure that supervision focuses on the most beneficial perspectives, we introduce a UCB-based strategy with a geometry-aware reward, which dynamically balances exploration and exploitation to identify informative viewpoints throughout training. Within this framework, we jointly optimize SDF geometry and view-dependent appearance under differentiable rendering, while periodically refining meshes to capture fine geometric details. Experiments demonstrate that our method achieves competitive results in both geometric accuracy and rendering quality.