Torsten Sattler

Affiliations

Chalmers University of Technology

papers

2,535

total citations

papers (28)

Mip-Splatting: Alias-free 3D Gaussian Splatting

CVPR 2024arXiv

627

citations

MonoSDF: Exploring Monocular Geometric Cues for Neural Implicit Surface Reconstruction

NEURIPS 2022arXiv

577

citations

Back to the Feature: Learning Robust Camera Localization From Pixels To Pose

CVPR 2021arXiv

296

citations

Patch2Pix: Epipolar-Guided Pixel-Level Correspondences

CVPR 2021arXiv

236

citations

Human POSEitioning System (HPS): 3D Human Pose Estimation and Self-Localization in Large Scenes From Body-Mounted Sensors

CVPR 2021arXiv

162

citations

Beyond Controlled Environments: 3D Camera Re-Localization in Changing Indoor Scenes

ECCV 2020arXiv

citations

Making Affine Correspondences Work in Camera Geometry Computation

ECCV 2020arXiv

citations

How Privacy-Preserving Are Line Clouds? Recovering Scene Details From 3D Lines

CVPR 2021arXiv

citations

Visual Localization Using Imperfect 3D Models From the Internet

CVPR 2023arXiv

citations

Infrastructure-based Multi-Camera Calibration using Radial Projections

ECCV 2020arXiv

citations

The Unreasonable Effectiveness of Pre-Trained Features for Camera Pose Refinement

CVPR 2024arXiv

citations

CrowdDriven: A New Challenging Dataset for Outdoor Visual Localization

ICCV 2021arXiv

citations

Calibrated and Partially Calibrated Semi-Generalized Homographies

ICCV 2021arXiv

citations

P1AC: Revisiting Absolute Pose From a Single Affine Correspondence

ICCV 2023arXiv

citations

Privacy-Preserving Representations Are Not Enough: Recovering Scene Content From Camera Poses

CVPR 2023arXiv

citations

RePoseD: Efficient Relative Pose Estimation With Known Depth Information

ICCV 2025arXiv

citations

Practical Solutions to the Relative Pose of Three Calibrated Cameras

CVPR 2025arXiv

citations

Gaussian Splatting Feature Fields for (Privacy-Preserving) Visual Localization

CVPR 2025arXiv

citations

Absolute Pose from One or Two Scaled and Oriented Features

CVPR 2024

citations

Handcrafted Outlier Detection Revisited

ECCV 2020

citations

SegLoc: Learning Segmentation-Based Representations for Privacy-Preserving Visual Localization

CVPR 2023

citations

Why Having 10,000 Parameters in Your Camera Model Is Better Than Twelve

CVPR 2020

citations

Objects Can Move: 3D Change Detection by Geometric Transformation Consistency

ECCV 2022

citations

Torsten Sattler

Affiliations

papers (28)

Mip-Splatting: Alias-free 3D Gaussian Splatting

MonoSDF: Exploring Monocular Geometric Cues for Neural Implicit Surface Reconstruction

Back to the Feature: Learning Robust Camera Localization From Pixels To Pose

Patch2Pix: Epipolar-Guided Pixel-Level Correspondences

Human POSEitioning System (HPS): 3D Human Pose Estimation and Self-Localization in Large Scenes From Body-Mounted Sensors

Deep Visual Geo-Localization Benchmark

ViewFormer: NeRF-Free Neural Rendering from Few Images Using Transformers

On the Limits of Pseudo Ground Truth in Visual Camera Re-Localisation

MeshLoc: Mesh-Based Visual Localization

Tetra-NeRF: Representing Neural Radiance Fields Using Tetrahedra

Beyond Controlled Environments: 3D Camera Re-Localization in Changing Indoor Scenes

Making Affine Correspondences Work in Camera Geometry Computation

How Privacy-Preserving Are Line Clouds? Recovering Scene Details From 3D Lines

Visual Localization Using Imperfect 3D Models From the Internet

Infrastructure-based Multi-Camera Calibration using Radial Projections

The Unreasonable Effectiveness of Pre-Trained Features for Camera Pose Refinement

CrowdDriven: A New Challenging Dataset for Outdoor Visual Localization

Calibrated and Partially Calibrated Semi-Generalized Homographies

P1AC: Revisiting Absolute Pose From a Single Affine Correspondence

Privacy-Preserving Representations Are Not Enough: Recovering Scene Content From Camera Poses

RePoseD: Efficient Relative Pose Estimation With Known Depth Information

Practical Solutions to the Relative Pose of Three Calibrated Cameras

Gaussian Splatting Feature Fields for (Privacy-Preserving) Visual Localization

Absolute Pose from One or Two Scaled and Oriented Features

Handcrafted Outlier Detection Revisited

SegLoc: Learning Segmentation-Based Representations for Privacy-Preserving Visual Localization

Why Having 10,000 Parameters in Your Camera Model Is Better Than Twelve

Objects Can Move: 3D Change Detection by Geometric Transformation Consistency

papers (28)

Mip-Splatting: Alias-free 3D Gaussian Splatting

MonoSDF: Exploring Monocular Geometric Cues for Neural Implicit Surface Reconstruction

Back to the Feature: Learning Robust Camera Localization From Pixels To Pose

Patch2Pix: Epipolar-Guided Pixel-Level Correspondences

Human POSEitioning System (HPS): 3D Human Pose Estimation and Self-Localization in Large Scenes From Body-Mounted Sensors

Deep Visual Geo-Localization Benchmark

ViewFormer: NeRF-Free Neural Rendering from Few Images Using Transformers

On the Limits of Pseudo Ground Truth in Visual Camera Re-Localisation

MeshLoc: Mesh-Based Visual Localization

Tetra-NeRF: Representing Neural Radiance Fields Using Tetrahedra

Beyond Controlled Environments: 3D Camera Re-Localization in Changing Indoor Scenes

Making Affine Correspondences Work in Camera Geometry Computation

How Privacy-Preserving Are Line Clouds? Recovering Scene Details From 3D Lines

Visual Localization Using Imperfect 3D Models From the Internet

Infrastructure-based Multi-Camera Calibration using Radial Projections

The Unreasonable Effectiveness of Pre-Trained Features for Camera Pose Refinement

CrowdDriven: A New Challenging Dataset for Outdoor Visual Localization

Calibrated and Partially Calibrated Semi-Generalized Homographies

P1AC: Revisiting Absolute Pose From a Single Affine Correspondence

Privacy-Preserving Representations Are Not Enough: Recovering Scene Content From Camera Poses

RePoseD: Efficient Relative Pose Estimation With Known Depth Information

Practical Solutions to the Relative Pose of Three Calibrated Cameras

Gaussian Splatting Feature Fields for (Privacy-Preserving) Visual Localization

Absolute Pose from One or Two Scaled and Oriented Features

Handcrafted Outlier Detection Revisited

SegLoc: Learning Segmentation-Based Representations for Privacy-Preserving Visual Localization

Why Having 10,000 Parameters in Your Camera Model Is Better Than Twelve

Objects Can Move: 3D Change Detection by Geometric Transformation Consistency