GaussianFormer: Scene as Gaussians for Vision-Based 3D Semantic Occupancy Prediction
- URL: http://arxiv.org/abs/2405.17429v1
- Date: Mon, 27 May 2024 17:59:51 GMT
- Title: GaussianFormer: Scene as Gaussians for Vision-Based 3D Semantic Occupancy Prediction
- Authors: Yuanhui Huang, Wenzhao Zheng, Yunpeng Zhang, Jie Zhou, Jiwen Lu,
- Abstract summary: 3D semantic occupancy prediction aims to obtain 3D fine-grained geometry and semantics of the surrounding scene.
We propose an object-centric representation to describe 3D scenes with sparse 3D semantic Gaussians.
GaussianFormer achieves comparable performance with state-of-the-art methods with only 17.8% - 24.8% of their memory consumption.
- Score: 70.65250036489128
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: 3D semantic occupancy prediction aims to obtain 3D fine-grained geometry and semantics of the surrounding scene and is an important task for the robustness of vision-centric autonomous driving. Most existing methods employ dense grids such as voxels as scene representations, which ignore the sparsity of occupancy and the diversity of object scales and thus lead to unbalanced allocation of resources. To address this, we propose an object-centric representation to describe 3D scenes with sparse 3D semantic Gaussians where each Gaussian represents a flexible region of interest and its semantic features. We aggregate information from images through the attention mechanism and iteratively refine the properties of 3D Gaussians including position, covariance, and semantics. We then propose an efficient Gaussian-to-voxel splatting method to generate 3D occupancy predictions, which only aggregates the neighboring Gaussians for a certain position. We conduct extensive experiments on the widely adopted nuScenes and KITTI-360 datasets. Experimental results demonstrate that GaussianFormer achieves comparable performance with state-of-the-art methods with only 17.8% - 24.8% of their memory consumption. Code is available at: https://github.com/huang-yh/GaussianFormer.
Related papers
- Adversarial Generation of Hierarchical Gaussians for 3D Generative Model [20.833116566243408]
In this paper, we exploit Gaussian as a 3D representation for 3D GANs by leveraging its efficient and explicit characteristics.
We introduce a generator architecture with a hierarchical multi-scale Gaussian representation that effectively regularizes the position and scale of generated Gaussians.
Experimental results demonstrate that ours achieves a significantly faster rendering speed (x100) compared to state-of-the-art 3D consistent GANs.
arXiv Detail & Related papers (2024-06-05T05:52:20Z) - CLIP-GS: CLIP-Informed Gaussian Splatting for Real-time and View-consistent 3D Semantic Understanding [32.76277160013881]
We present CLIP-GS, which integrates semantics from Contrastive Language-Image Pre-Training (CLIP) into Gaussian Splatting.
SAC exploits the inherent unified semantics within objects to learn compact yet effective semantic representations of 3D Gaussians.
We also introduce a 3D Coherent Self-training (3DCS) strategy, resorting to the multi-view consistency originated from the 3D model.
arXiv Detail & Related papers (2024-04-22T15:01:32Z) - Semantic Gaussians: Open-Vocabulary Scene Understanding with 3D Gaussian Splatting [27.974762304763694]
Open-vocabulary 3D scene understanding presents a significant challenge in computer vision.
We introduce SemanticGaussians, a novel open-vocabulary scene understanding approach based on 3D Gaussian Splatting.
Our approach attains a 4.2% mIoU and 4.0%mAcc improvement over prior open-vocabulary scene understanding counterparts.
arXiv Detail & Related papers (2024-03-22T21:28:19Z) - HUGS: Holistic Urban 3D Scene Understanding via Gaussian Splatting [53.6394928681237]
holistic understanding of urban scenes based on RGB images is a challenging yet important problem.
Our main idea involves the joint optimization of geometry, appearance, semantics, and motion using a combination of static and dynamic 3D Gaussians.
Our approach offers the ability to render new viewpoints in real-time, yielding 2D and 3D semantic information with high accuracy.
arXiv Detail & Related papers (2024-03-19T13:39:05Z) - Compact 3D Scene Representation via Self-Organizing Gaussian Grids [10.816451552362823]
3D Gaussian Splatting has recently emerged as a highly promising technique for modeling of static 3D scenes.
We introduce a compact scene representation organizing the parameters of 3DGS into a 2D grid with local homogeneity.
Our method achieves a reduction factor of 17x to 42x in size for complex scenes with no increase in training time.
arXiv Detail & Related papers (2023-12-19T20:18:29Z) - GAvatar: Animatable 3D Gaussian Avatars with Implicit Mesh Learning [60.33970027554299]
Gaussian splatting has emerged as a powerful 3D representation that harnesses the advantages of both explicit (mesh) and implicit (NeRF) 3D representations.
In this paper, we seek to leverage Gaussian splatting to generate realistic animatable avatars from textual descriptions.
Our proposed method, GAvatar, enables the large-scale generation of diverse animatable avatars using only text prompts.
arXiv Detail & Related papers (2023-12-18T18:59:12Z) - GIR: 3D Gaussian Inverse Rendering for Relightable Scene Factorization [76.52007427483396]
GIR is a 3D Gaussian Inverse Rendering method for relightable scene factorization.
Our method utilizes 3D Gaussians to estimate the material properties, illumination, and geometry of an object from multi-view images.
arXiv Detail & Related papers (2023-12-08T16:05:15Z) - Gaussian Grouping: Segment and Edit Anything in 3D Scenes [65.49196142146292]
We propose Gaussian Grouping, which extends Gaussian Splatting to jointly reconstruct and segment anything in open-world 3D scenes.
Compared to the implicit NeRF representation, we show that the grouped 3D Gaussians can reconstruct, segment and edit anything in 3D with high visual quality, fine granularity and efficiency.
arXiv Detail & Related papers (2023-12-01T17:09:31Z) - GS-SLAM: Dense Visual SLAM with 3D Gaussian Splatting [51.96353586773191]
We introduce textbfGS-SLAM that first utilizes 3D Gaussian representation in the Simultaneous Localization and Mapping system.
Our method utilizes a real-time differentiable splatting rendering pipeline that offers significant speedup to map optimization and RGB-D rendering.
Our method achieves competitive performance compared with existing state-of-the-art real-time methods on the Replica, TUM-RGBD datasets.
arXiv Detail & Related papers (2023-11-20T12:08:23Z)
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.