CATSplat: Context-Aware Transformer with Spatial Guidance for Generalizable 3D Gaussian Splatting from A Single-View Image
- URL: http://arxiv.org/abs/2412.12906v2
- Date: Mon, 03 Feb 2025 08:06:36 GMT
- Title: CATSplat: Context-Aware Transformer with Spatial Guidance for Generalizable 3D Gaussian Splatting from A Single-View Image
- Authors: Wonseok Roh, Hwanhee Jung, Jong Wook Kim, Seunggwan Lee, Innfarn Yoo, Andreas Lugmayr, Seunggeun Chi, Karthik Ramani, Sangpil Kim,
- Abstract summary: We introduce CATSplat, a novel generalizable transformer-based framework for single-view 3D scene reconstruction.
By incorporating scene-specific contextual details from text embeddings through cross-attention, we pave the way for context-aware reconstruction.
Experiments on large-scale datasets demonstrate the state-of-the-art performance of CATSplat in single-view 3D scene reconstruction.
- Score: 18.445769892372528
- License:
- Abstract: Recently, generalizable feed-forward methods based on 3D Gaussian Splatting have gained significant attention for their potential to reconstruct 3D scenes using finite resources. These approaches create a 3D radiance field, parameterized by per-pixel 3D Gaussian primitives, from just a few images in a single forward pass. However, unlike multi-view methods that benefit from cross-view correspondences, 3D scene reconstruction with a single-view image remains an underexplored area. In this work, we introduce CATSplat, a novel generalizable transformer-based framework designed to break through the inherent constraints in monocular settings. First, we propose leveraging textual guidance from a visual-language model to complement insufficient information from a single image. By incorporating scene-specific contextual details from text embeddings through cross-attention, we pave the way for context-aware 3D scene reconstruction beyond relying solely on visual cues. Moreover, we advocate utilizing spatial guidance from 3D point features toward comprehensive geometric understanding under single-view settings. With 3D priors, image features can capture rich structural insights for predicting 3D Gaussians without multi-view techniques. Extensive experiments on large-scale datasets demonstrate the state-of-the-art performance of CATSplat in single-view 3D scene reconstruction with high-quality novel view synthesis.
Related papers
- FreeSplat: Generalizable 3D Gaussian Splatting Towards Free-View Synthesis of Indoor Scenes [50.534213038479926]
FreeSplat is capable of reconstructing geometrically consistent 3D scenes from long sequence input towards free-view synthesis.
We propose a simple but effective free-view training strategy that ensures robust view synthesis across broader view range regardless of the number of views.
arXiv Detail & Related papers (2024-05-28T08:40:14Z) - Denoising Diffusion via Image-Based Rendering [54.20828696348574]
We introduce the first diffusion model able to perform fast, detailed reconstruction and generation of real-world 3D scenes.
First, we introduce a new neural scene representation, IB-planes, that can efficiently and accurately represent large 3D scenes.
Second, we propose a denoising-diffusion framework to learn a prior over this novel 3D scene representation, using only 2D images.
arXiv Detail & Related papers (2024-02-05T19:00:45Z) - SceneWiz3D: Towards Text-guided 3D Scene Composition [134.71933134180782]
Existing approaches either leverage large text-to-image models to optimize a 3D representation or train 3D generators on object-centric datasets.
We introduce SceneWiz3D, a novel approach to synthesize high-fidelity 3D scenes from text.
arXiv Detail & Related papers (2023-12-13T18:59:30Z) - Single-view 3D Scene Reconstruction with High-fidelity Shape and Texture [47.44029968307207]
We propose a novel framework for simultaneous high-fidelity recovery of object shapes and textures from single-view images.
Our approach utilizes the proposed Single-view neural implicit Shape and Radiance field (SSR) representations to leverage both explicit 3D shape supervision and volume rendering.
A distinctive feature of our framework is its ability to generate fine-grained textured meshes while seamlessly integrating rendering capabilities into the single-view 3D reconstruction model.
arXiv Detail & Related papers (2023-11-01T11:46:15Z) - Learning to Render Novel Views from Wide-Baseline Stereo Pairs [26.528667940013598]
We introduce a method for novel view synthesis given only a single wide-baseline stereo image pair.
Existing approaches to novel view synthesis from sparse observations fail due to recovering incorrect 3D geometry.
We propose an efficient, image-space epipolar line sampling scheme to assemble image features for a target ray.
arXiv Detail & Related papers (2023-04-17T17:40:52Z) - 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) - Vision Transformer for NeRF-Based View Synthesis from a Single Input
Image [49.956005709863355]
We propose to leverage both the global and local features to form an expressive 3D representation.
To synthesize a novel view, we train a multilayer perceptron (MLP) network conditioned on the learned 3D representation to perform volume rendering.
Our method can render novel views from only a single input image and generalize across multiple object categories using a single model.
arXiv Detail & Related papers (2022-07-12T17:52:04Z)
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.