Fredrik Kahl
Fredrik Kahl är professor och ledare för datorseendegruppen. För mer information, se https://fredkahl.github.io/
Visar 69 publikationer
Learning Structure-from-Motion with Graph Attention Networks
Privacy-Preserving Representations are not Enough: Recovering Scene Content from Camera Poses
Rigidity Preserving Image Transformations and Equivariance in Perspective
Long-Term Visual Localization Revisited
ZZ-Net: A Universal Rotation Equivariant Architecture for 2D Point Clouds
Azimuthal Rotational Equivariance in Spherical Convolutional Neural Networks
A case for using rotation invariant features in state of the art feature matchers
DoubleMatch: Improving Semi-Supervised Learning with Self-Supervision
On the Tightness of Semidefinite Relaxations for Rotation Estimation
A quasiconvex formulation for radial cameras
CrowdDriven: A New Challenging Dataset for Outdoor Visual Localization
Rotation Averaging with the Chordal Distance: Global Minimizers and Strong Duality
How privacy-preserving are line clouds? Recovering scene details from 3D lines
Back to the Feature: Learning Robust Camera Localization from Pixels to Pose
Global Optimality for Point Set Registration Using Semidefinite Programming
Single-Image Depth Prediction Makes Feature Matching Easier
Pose Proposal Critic: Robust Pose Refinement by Learning Reprojection Errors
Combining Depth Fusion and Photometric Stereo for Fine-Detailed 3D Models
Shape-aware label fusion for multi-atlas frameworks
A cross-season correspondence dataset for robust semantic segmentation
A Deep Learning Approach to MR-less Spatial Normalization for Tau PET Images
Revisiting Deep Structured Models for Pixel-Level Labeling with Gradient-Based Inference
Semantic Match Consistency for Long-Term Visual Localization
Robust Abdominal Organ Segmentation Using Regional Convolutional Neural Networks
Multiresolution Search of the Rigid Motion Space for Intensity-Based Registration
Rotation Averaging and Strong Duality
Benchmarking 6DOF Outdoor Visual Localization in Changing Conditions
Multiatlas Segmentation Using Robust Feature-Based Registration
Robust abdominal organ segmentation using regional convolutional neural networks
Max-margin learning of deep structured models for semantic segmentation
City-scale localization for cameras with known vertical direction
Long-term 3D Localization and Pose from Semantic Labellings
Optimal relative pose with unknown correspondences
Shape-aware multi-atlas segmentation
A Simple Method for Subspace Estimation with Corrupted Columns
Optimal Gradient Encoding Schemes for Diffusion Tensor and Kurtosis Imaging
Outlier rejection for absolute pose estimation with known orientation
Robust online 3D reconstruction combining a depth sensor and sparse feature points
Efficient algorithms for robust estimation of relative translation
Image Fusion of Reconstructed Digital Tomosynthesis Volumes From a Frontal and a Lateral Acquisition
Automated segmentation of the skeleton in PET/CT scans
Deepseg: Abdominal Organ Segmentation Using Deep Convolutional Neural Networks
Globally optimal rigid intensity based registration: A fast fourier domain approach
Überatlas: Fast and robust registration for multi-atlas segmentation
Good Features for Reliable Registration in Multi-Atlas Segmentation
Tractable Algorithms for Robust Model Estimation
Improving Robustness for Inter-Subject Medical Image Registration Using a Feature-Based Approach
Überatlas: Robust Speed-Up of Feature-Based Registration and Multi-Atlas Segmentation
Shortest Paths with Higher-Order Regularization
A Combinatorial Approach to L1-Matrix Factorization
Robust Camera Tracking by Combining Color and Depth Measurements
Accurate Localization and Pose Estimation for Large 3D Models
Tractable and Reliable Registration of 2D Point Sets
Fast and Reliable Two-View Translation Estimation
Minimal solvers for relative pose with a single unknown radial distortion
Rank Minimization with Structured Data Patterns
Shortest Paths with Curvature and Torsion
Ladda ner publikationslistor
Du kan ladda ner denna lista till din dator.
Filtrera och ladda ner publikationslista
Som inloggad användare hittar du ytterligare funktioner i MyResearch.
Du kan även exportera direkt till Zotero eller Mendeley genom webbläsarplugins. Dessa hittar du här:
Zotero Connector
Mendeley Web Importer
Tjänsten SwePub erbjuder uttag av Researchs listor i andra format, till exempel kan du få uttag av publikationer enligt Harvard och Oxford i .RIS, BibTex och RefWorks-format.
Visar 17 forskningsprojekt
Deep learning-based out of distribution detection of computed tomography images
Semiövervakad inlärning för medicinsk bildanalys
Robust och precis semi-övervakad inlärning
Deep Learning för att extrahera trädstrukturer i medicinska bilder
Automatic data analysis within non-destructive evaluation (ADA-NDE)
Optimeringsmetoder med prestandagarantier för maskininlärningsmetoder
INNER: information theory of deep neural networks
Deep learning för 3D-igenkänning
Datorseende för gradering av hudsjukdomar
Integrering av geometri och semantik i datorseende
Semantisk kartering & visuell navigering för smarta robotar
Robusta och tillförlitliga metoder för modellanpassning i datorseende