Results 21 to 30 of about 2,993 (218)

Deep Projective 3D Semantic Segmentation

, 2017
Semantic segmentation of 3D point clouds is a challenging problem with numerous real-world applications. While deep learning has revolutionized the field of image semantic segmentation, its impact on point cloud data has been limited so far.
AE Johnson, JA Montoya-Zegarra, M Kazhdan, R Szeliski, S Gupta, S Salti, T Hackel   +6 more
core   +1 more source

GAGS: Gradient-Guided Adaptive Gaussian Splatting for Efficient and Geometry-Regularized Surface Reconstruction [PDF]

ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
While 3D Gaussian Splatting (3DGS) has advanced surface reconstruction, existing implementations face critical challenges in memory efficiency and geometric fidelity. Current approaches like PGSR generate excessive Gaussian primitives due to uncontrolled
Y. Hou, T. Wang, X. Wang, Z. Zhan
doaj   +1 more source

Mip-Splatting: Alias-Free 3D Gaussian Splatting

2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Project page: https://niujinshuchong.github.io/mip-splatting/
Yu, Zehao, Chen, Anpei, Huang, Binbin, Sattler, Torsten, Geiger, Andreas   +4 more
openaire   +2 more sources

Learning Sparse High Dimensional Filters: Image Filtering, Dense CRFs and Bilateral Neural Networks

, 2015
Bilateral filters have wide spread use due to their edge-preserving properties. The common use case is to manually choose a parametric filter type, usually a Gaussian filter.
Gehler, Peter V., Jampani, Varun, Kiefel, Martin   +2 more
core   +1 more source

Frequency Analysis of Gradient Estimators in Volume Rendering [PDF]

, 1995
Gradient information is used in volume rendering to classify and color samples along a ray. In this paper, we present an analysis of the theoretically ideal gradient estimator and compare it to some commonly used gradient estimators.
Bentum, Mark J., Lichtenbelt, Barthold B.A., Malzbender, Tom   +2 more
core   +4 more sources

HoloGaussian Digital Twin: Reconstructing 3D Scenes with Gaussian Splatting for Tabletop Hologram Visualization of Real Environments

Remote Sensing
Several studies have explored the use of hologram technology in architecture and urban design, demonstrating its feasibility. Holograms can represent 3D spatial data and offer an immersive experience, potentially replacing traditional methods such as ...
Tam Le Phuc Do, Jinwon Choi, Viet Quoc Le, Philippe Gentet, Leehwan Hwang, Seunghyun Lee   +5 more
doaj   +1 more source

Modeling Uncertainty for Gaussian Splatting

IEEE Transactions on Neural Networks and Learning Systems
We present Stochastic Gaussian Splatting (SGS): the first framework for uncertainty estimation using Gaussian Splatting (GS). GS recently advanced the novel-view synthesis field by achieving impressive reconstruction quality at a fraction of the computational cost of Neural Radiance Fields (NeRF).
Aira Luca Savant, Valsesia Diego, Magli Enrico   +2 more
openaire   +4 more sources

Efficient Relaxations for Dense CRFs with Sparse Higher Order Potentials

, 2018
Dense conditional random fields (CRFs) have become a popular framework for modelling several problems in computer vision such as stereo correspondence and multi-class semantic segmentation.
Ajanthan, Thalaiyasingam, Bunel, Rudy, Desmaison, Alban, Joy, Thomas, Kohli, Pushmeet, Kumar, M. Pawan, Salzmann, Mathieu, Torr, Philip H. S.   +7 more
core   +1 more source

Advances in Real-Time Reconstruction of Deformable Tissues for Endoscopic Surgery: A Comprehensive Review of Gaussian Splatting Techniques and Applications

IEEE Access
Endoscopic imaging has become an indispensable tool in modern medical practice, enabling minimally invasive diagnosis and treatment. A crucial aspect of endoscopic procedures is the accurate reconstruction of the observed scene, which facilitates spatial
Di Ding, Haoyu Wang, Tianliang Yao, Rong Luo, Xusen Sun   +4 more
doaj   +1 more source

3D Gaussian Splatting for Enhanced Documentation of Cultural Artifacts [PDF]

ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
3D Gaussian splatting has shown promise for high-fidelity 3D modeling in cultural heritage documentation. However, applying 3DGS to cultural artifacts faces challenges, including image alignment errors in dynamic capture environments and background ...
J. Lee, S. Park, J. Min, J. Choi
doaj   +1 more source