GaussRender: Learning 3D Occupancy with Gaussian Rendering

• URL: http://arxiv.org/abs/2502.05040v1
• Date: Fri, 07 Feb 2025 16:07:51 GMT
• Title: GaussRender: Learning 3D Occupancy with Gaussian Rendering
• Authors: Loick Chambon, Eloi Zablocki, Alexandre Boulch, Mickael Chen, Matthieu Cord,
• Abstract summary: GaussRender is a plug-and-play 3D-to-2D reprojection loss that enhances voxel-based supervision. Our method projects 3D voxel representations into arbitrary 2D perspectives and leverages Gaussian splatting as an efficient, differentiable rendering proxy of voxels.
• Score: 84.60008381280286
• License:
• Abstract: Understanding the 3D geometry and semantics of driving scenes is critical for developing of safe autonomous vehicles. While 3D occupancy models are typically trained using voxel-based supervision with standard losses (e.g., cross-entropy, Lovasz, dice), these approaches treat voxel predictions independently, neglecting their spatial relationships. In this paper, we propose GaussRender, a plug-and-play 3D-to-2D reprojection loss that enhances voxel-based supervision. Our method projects 3D voxel representations into arbitrary 2D perspectives and leverages Gaussian splatting as an efficient, differentiable rendering proxy of voxels, introducing spatial dependencies across projected elements. This approach improves semantic and geometric consistency, handles occlusions more efficiently, and requires no architectural modifications. Extensive experiments on multiple benchmarks (SurroundOcc-nuScenes, Occ3D-nuScenes, SSCBench-KITTI360) demonstrate consistent performance gains across various 3D occupancy models (TPVFormer, SurroundOcc, Symphonies), highlighting the robustness and versatility of our framework. The code is available at https://github.com/valeoai/GaussRender.

Related papers

• Robust 3D Semantic Occupancy Prediction with Calibration-free Spatial Transformation [32.50849425431012]
  For autonomous cars equipped with multi-camera and LiDAR, it is critical to aggregate multi-sensor information into a unified 3D space for accurate and robust predictions. Recent methods are mainly built on the 2D-to-3D transformation that relies on sensor calibration to project the 2D image information into the 3D space. In this work, we propose a calibration-free spatial transformation based on vanilla attention to implicitly model the spatial correspondence.
  arXiv  Detail & Related papers  (2024-11-19T02:40:42Z)
• Neural Signed Distance Function Inference through Splatting 3D Gaussians Pulled on Zero-Level Set [49.780302894956776]
  It is vital to infer a signed distance function (SDF) in multi-view based surface reconstruction. We propose a method that seamlessly merge 3DGS with the learning of neural SDFs. Our numerical and visual comparisons show our superiority over the state-of-the-art results on the widely used benchmarks.
  arXiv  Detail & Related papers  (2024-10-18T05:48:06Z)
• Enhancing Single Image to 3D Generation using Gaussian Splatting and Hybrid Diffusion Priors [17.544733016978928]
  3D object generation from a single image involves estimating the full 3D geometry and texture of unseen views from an unposed RGB image captured in the wild. Recent advancements in 3D object generation have introduced techniques that reconstruct an object's 3D shape and texture. We propose bridging the gap between 2D and 3D diffusion models to address this limitation.
  arXiv  Detail & Related papers  (2024-10-12T10:14:11Z)
• GSD: View-Guided Gaussian Splatting Diffusion for 3D Reconstruction [52.04103235260539]
  We present a diffusion model approach based on Gaussian Splatting representation for 3D object reconstruction from a single view. The model learns to generate 3D objects represented by sets of GS ellipsoids. The final reconstructed objects explicitly come with high-quality 3D structure and texture, and can be efficiently rendered in arbitrary views.
  arXiv  Detail & Related papers  (2024-07-05T03:43:08Z)
• Any2Point: Empowering Any-modality Large Models for Efficient 3D Understanding [83.63231467746598]
  We introduce Any2Point, a parameter-efficient method to empower any-modality large models (vision, language, audio) for 3D understanding. We propose a 3D-to-any (1D or 2D) virtual projection strategy that correlates the input 3D points to the original 1D or 2D positions within the source modality.
  arXiv  Detail & Related papers  (2024-04-11T17:59:45Z)
• Semantic Gaussians: Open-Vocabulary Scene Understanding with 3D Gaussian Splatting [27.974762304763694]
  We introduce Semantic Gaussians, a novel open-vocabulary scene understanding approach based on 3D Gaussian Splatting. Unlike existing methods, we design a versatile projection approach that maps various 2D semantic features into a novel semantic component of 3D Gaussians. We build a 3D semantic network that directly predicts the semantic component from raw 3D Gaussians for fast inference.
  arXiv  Detail & Related papers  (2024-03-22T21:28:19Z)
• Volumetric Environment Representation for Vision-Language Navigation [66.04379819772764]
  Vision-language navigation (VLN) requires an agent to navigate through a 3D environment based on visual observations and natural language instructions. We introduce a Volumetric Environment Representation (VER), which voxelizes the physical world into structured 3D cells. VER predicts 3D occupancy, 3D room layout, and 3D bounding boxes jointly.
  arXiv  Detail & Related papers  (2024-03-21T06:14:46Z)
• PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm [114.47216525866435]
  We introduce a novel universal 3D pre-training framework designed to facilitate the acquisition of efficient 3D representation. For the first time, PonderV2 achieves state-of-the-art performance on 11 indoor and outdoor benchmarks, implying its effectiveness.
  arXiv  Detail & Related papers  (2023-10-12T17:59:57Z)
• RenderOcc: Vision-Centric 3D Occupancy Prediction with 2D Rendering Supervision [36.15913507034939]
  We present RenderOcc, a novel paradigm for training 3D occupancy models only using 2D labels. Specifically, we extract a NeRF-style 3D volume representation from multi-view images. We employ volume rendering techniques to establish 2D renderings, thus enabling direct 3D supervision from 2D semantics and depth labels.
  arXiv  Detail & Related papers  (2023-09-18T06:08:15Z)

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.