Experiments show that SpaRV keeps the geometry consistent away from the captured trajectory and sharpens textures by optimizing the camera parameters, which means that SpaRV keeps the geometry consistent away from the captured trajectory and sharpens textures by optimizing the camera parameters.
Abstract
Robotic simulation, navigation, and inspection need photorealistic and geometrically accurate maps. Radiance fields achieve photorealism, but building them from robot data raises three challenges. A robot's trajectory serves its task, not reconstruction, so images from few viewpoints leave the geometry under-constrained. Residual errors in camera poses and intrinsics blur the map. The map's GPU memory grows with scene size, which limits its level of detail in large scenes. We present \textbf{SpaRV}, a \textbf{Spa}rse \textbf{R}adiance \textbf{V}oxel map rendered by differentiable ray marching, which lets irregular LiDAR scans supervise it alongside images. LiDAR depth and occupancy anchor the under-constrained geometry, and normal and smoothness regularizers keep it smooth between LiDAR points. To correct camera errors, SpaRV alternates map training with photometric optimization of the camera poses and focal lengths. To bound GPU memory in training and rendering, SpaRV holds one submap at a time at full resolution and the rest at coarser octree levels. Experiments show that SpaRV keeps the geometry consistent away from the captured trajectory and sharpens textures by optimizing the camera parameters. On one 24\,GB GPU, SpaRV reconstructs a university campus captured in 63{,}582 images and simulates other cameras and LiDARs. Our code will be released at \url{https://github.com/hku-mars/SpaRV}.
Vision-Based Navigation (VBN) must operate over surfaces whose appearance is dictatedby their material optical properties. Because flight-representative imagery with 3D groundtruth is scarce, learning-based VBN relies on simulation — and a simulator is only asfaithful as the reflectance model it uses. This paper follow...
Clémentine Grethen, G. Morin, Simone Gasparini et al.· Papers of ESA GNC-ICATT 2023· 0 citations
Cameras and LiDARs (Light Detection and Ranging) provide complementary observations of the same scene: cameras capture rich appearance cues while LiDAR sensors provide accurate geometric measurements. In this work, we address the problem of fine image-to-point cloud registration from a single image and a roughly estima...
Ugo Insalaco, Jean-Emmanuel Deschaud, Pierre Dellenbach· 0 citations
Camera localization in a prior LiDAR map provides a persistent geometric reference for long-term robotic navigation, yet remains challenging because of the substantial modality gap between camera images and point-cloud maps. We present a unified localization framework that makes the LiDAR map visually addressable rathe...
RangeNav is presented, a compact (183k parameters) LiDAR range-image encoder pretrained entirely on synthetic data that demonstrates robust navigation through dense crowds, low-light settings, and construction zones where limited-field-of-view depth policies struggle.
Samuel Foo Enze, Manthan Patel, Fan Yang et al.· 0 citations
Monocular outdoor navigation requires robust camera motion estimation and a dense, navigable map, where 3D Gaussian Splatting SLAM serves this task through an explicit differentiable parameterization and closed-form projection. However, typically, the geometry is recovered by associating 2D pixels with 3D points, so a...
Ziang Guo, Didar Seyidov, Yi-Xiao Zhou et al.· 0 citations
Accurate registration between aerial imagery and LiDAR point clouds is fundamental to building-level analysis and urban modeling. Although coarse alignment can be achieved through geo-referencing or sensor calibration, residual misalignment often remains and affects the reliable interpretation of roof structures. To ad...
Qi-Peng Mei, D. Bulatov, D. Iwaszczuk· The International Archives o...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.