GTR: Improving Large 3D Reconstruction Models through Geometry and Texture Refinement

• URL: http://arxiv.org/abs/2406.05649v2
• Date: Thu, 13 Jun 2024 18:18:15 GMT
• Title: GTR: Improving Large 3D Reconstruction Models through Geometry and Texture Refinement
• Authors: Peiye Zhuang, Songfang Han, Chaoyang Wang, Aliaksandr Siarohin, Jiaxu Zou, Michael Vasilkovsky, Vladislav Shakhrai, Sergey Korolev, Sergey Tulyakov, Hsin-Ying Lee,
• Abstract summary: We propose a novel approach for 3D mesh reconstruction from multi-view images. Our method takes inspiration from large reconstruction models that use a transformer-based triplane generator and a Neural Radiance Field (NeRF) model trained on multi-view images.
• Score: 51.97726804507328
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We propose a novel approach for 3D mesh reconstruction from multi-view images. Our method takes inspiration from large reconstruction models like LRM that use a transformer-based triplane generator and a Neural Radiance Field (NeRF) model trained on multi-view images. However, in our method, we introduce several important modifications that allow us to significantly enhance 3D reconstruction quality. First of all, we examine the original LRM architecture and find several shortcomings. Subsequently, we introduce respective modifications to the LRM architecture, which lead to improved multi-view image representation and more computationally efficient training. Second, in order to improve geometry reconstruction and enable supervision at full image resolution, we extract meshes from the NeRF field in a differentiable manner and fine-tune the NeRF model through mesh rendering. These modifications allow us to achieve state-of-the-art performance on both 2D and 3D evaluation metrics, such as a PSNR of 28.67 on Google Scanned Objects (GSO) dataset. Despite these superior results, our feed-forward model still struggles to reconstruct complex textures, such as text and portraits on assets. To address this, we introduce a lightweight per-instance texture refinement procedure. This procedure fine-tunes the triplane representation and the NeRF color estimation model on the mesh surface using the input multi-view images in just 4 seconds. This refinement improves the PSNR to 29.79 and achieves faithful reconstruction of complex textures, such as text. Additionally, our approach enables various downstream applications, including text- or image-to-3D generation.

Related papers

• $R^2$-Mesh: Reinforcement Learning Powered Mesh Reconstruction via Geometry and Appearance Refinement [5.810659946867557]
  Mesh reconstruction based on Neural Radiance Fields (NeRF) is popular in a variety of applications such as computer graphics, virtual reality, and medical imaging. We propose a novel algorithm that progressively generates and optimize meshes from multi-view images. Our method delivers highly competitive and robust performance in both mesh rendering quality and geometric quality.
  arXiv  Detail & Related papers  (2024-08-19T16:33:17Z)
• From Flat to Spatial: Comparison of 4 methods constructing 3D, 2 and 1/2D Models from 2D Plans with neural networks [0.0]
  The conversion of single images into 2 and 1/2D and 3D meshes is a promising technology that enhances design visualization and efficiency. This paper evaluates four innovative methods: "One-2-3-45," " CRM: Single Image to 3D Textured Mesh with Convolutional Reconstruction Model," "Instant Mesh," and "Image-to-Mesh"
  arXiv  Detail & Related papers  (2024-07-29T13:01:20Z)
• MaRINeR: Enhancing Novel Views by Matching Rendered Images with Nearby References [49.71130133080821]
  MaRINeR is a refinement method that leverages information of a nearby mapping image to improve the rendering of a target viewpoint. We show improved renderings in quantitative metrics and qualitative examples from both explicit and implicit scene representations.
  arXiv  Detail & Related papers  (2024-07-18T17:50:03Z)
• Fine-Grained Multi-View Hand Reconstruction Using Inverse Rendering [11.228453237603834]
  We present a novel fine-grained multi-view hand mesh reconstruction method that leverages inverse rendering to restore hand poses and intricate details. We also introduce a novel Hand Albedo and Mesh (HAM) optimization module to refine both the hand mesh and textures. Our proposed approach outperforms the state-of-the-art methods on both reconstruction accuracy and rendering quality.
  arXiv  Detail & Related papers  (2024-07-08T07:28:24Z)
• GeoLRM: Geometry-Aware Large Reconstruction Model for High-Quality 3D Gaussian Generation [65.33726478659304]
  We introduce the Geometry-Aware Large Reconstruction Model (GeoLRM), an approach which can predict high-quality assets with 512k Gaussians and 21 input images in only 11 GB GPU memory. Previous works neglect the inherent sparsity of 3D structure and do not utilize explicit geometric relationships between 3D and 2D images. GeoLRM tackles these issues by incorporating a novel 3D-aware transformer structure that directly processes 3D points and uses deformable cross-attention mechanisms.
  arXiv  Detail & Related papers  (2024-06-21T17:49:31Z)
• CRM: Single Image to 3D Textured Mesh with Convolutional Reconstruction Model [37.75256020559125]
  We present a high-fidelity feed-forward single image-to-3D generative model. We highlight the necessity of integrating geometric priors into network design. Our model delivers a high-fidelity textured mesh from an image in just 10 seconds, without any test-time optimization.
  arXiv  Detail & Related papers  (2024-03-08T04:25:29Z)
• One-2-3-45: Any Single Image to 3D Mesh in 45 Seconds without Per-Shape Optimization [30.951405623906258]
  Single image 3D reconstruction is an important but challenging task that requires extensive knowledge of our natural world. We propose a novel method that takes a single image of any object as input and generates a full 360-degree 3D textured mesh in a single feed-forward pass.
  arXiv  Detail & Related papers  (2023-06-29T13:28:16Z)
• Delicate Textured Mesh Recovery from NeRF via Adaptive Surface Refinement [78.48648360358193]
  We present a novel framework that generates textured surface meshes from images. Our approach begins by efficiently initializing the geometry and view-dependency appearance with a NeRF. We jointly refine the appearance with geometry and bake it into texture images for real-time rendering.
  arXiv  Detail & Related papers  (2023-03-03T17:14:44Z)
• High-fidelity 3D GAN Inversion by Pseudo-multi-view Optimization [51.878078860524795]
  We present a high-fidelity 3D generative adversarial network (GAN) inversion framework that can synthesize photo-realistic novel views. Our approach enables high-fidelity 3D rendering from a single image, which is promising for various applications of AI-generated 3D content.
  arXiv  Detail & Related papers  (2022-11-28T18:59:52Z)
• Fast-GANFIT: Generative Adversarial Network for High Fidelity 3D Face Reconstruction [76.1612334630256]
  We harness the power of Generative Adversarial Networks (GANs) and Deep Convolutional Neural Networks (DCNNs) to reconstruct the facial texture and shape from single images. We demonstrate excellent results in photorealistic and identity preserving 3D face reconstructions and achieve for the first time, facial texture reconstruction with high-frequency details.
  arXiv  Detail & Related papers  (2021-05-16T16:35:44Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.