Search Results for author: Fredrik Kahl

Found 29 papers, 9 papers with code

Improving Open-Set Semi-Supervised Learning with Self-Supervision

no code implementations24 Jan 2023 Erik Wallin, Lennart Svensson, Fredrik Kahl, Lars Hammarstrand

In contrast, we propose an OSSL framework that facilitates learning from all unlabeled data through self-supervision.

Open Set Learning

In Search of Projectively Equivariant Neural Networks

no code implementations29 Sep 2022 Georg Bökman, Axel Flinth, Fredrik Kahl

Equivariance of linear neural network layers is well studied.

A case for using rotation invariant features in state of the art feature matchers

1 code implementation21 Apr 2022 Georg Bökman, Fredrik Kahl

The aim of this paper is to demonstrate that a state of the art feature matcher (LoFTR) can be made more robust to rotations by simply replacing the backbone CNN with a steerable CNN which is equivariant to translations and image rotations.

Rigidity Preserving Image Transformations and Equivariance in Perspective

no code implementations31 Jan 2022 Lucas Brynte, Georg Bökman, Axel Flinth, Fredrik Kahl

We characterize the class of image plane transformations which realize rigid camera motions and call these transformations `rigidity preserving'.

6D Pose Estimation using RGB Inductive Bias +1

CrowdDriven: A New Challenging Dataset for Outdoor Visual Localization

no code implementations ICCV 2021 Ara Jafarzadeh, Manuel Lopez Antequera, Pau Gargallo, Yubin Kuang, Carl Toft, Fredrik Kahl, Torsten Sattler

Visual localization is the problem of estimating the position and orientation from which a given image (or a sequence of images) is taken in a known scene.

Benchmarking Self-Driving Cars +1

A Quasiconvex Formulation for Radial Cameras

no code implementations CVPR 2021 Carl Olsson, Viktor Larsson, Fredrik Kahl

In this paper we study structure from motion problems for 1D radial cameras.

Back to the Feature: Learning Robust Camera Localization from Pixels to Pose

1 code implementation CVPR 2021 Paul-Edouard Sarlin, Ajaykumar Unagar, Måns Larsson, Hugo Germain, Carl Toft, Viktor Larsson, Marc Pollefeys, Vincent Lepetit, Lars Hammarstrand, Fredrik Kahl, Torsten Sattler

In this paper, we go Back to the Feature: we argue that deep networks should focus on learning robust and invariant visual features, while the geometric estimation should be left to principled algorithms.

Camera Localization Metric Learning +1

How Privacy-Preserving are Line Clouds? Recovering Scene Details from 3D Lines

1 code implementation CVPR 2021 Kunal Chelani, Fredrik Kahl, Torsten Sattler

To address the resulting potential privacy risks for user-generated content, it was recently proposed to lift point clouds to line clouds by replacing 3D points by randomly oriented 3D lines passing through these points.

Pose Estimation Privacy Preserving +2

On the Tightness of Semidefinite Relaxations for Rotation Estimation

no code implementations6 Jan 2021 Lucas Brynte, Viktor Larsson, José Pedro Iglesias, Carl Olsson, Fredrik Kahl

In studying the empirical performance we note that there are few failure cases reported in the literature, in particular for estimation problems with a single rotation, motivating us to gain further theoretical understanding.

Azimuthal Rotational Equivariance in Spherical CNNs

no code implementations1 Jan 2021 Carl Toft, Georg Bökman, Fredrik Kahl

In this work, we analyze linear operators from $L^2(S^2) \rightarrow L^2(S^2)$ which are equivariant to azimuthal rotations, that is, rotations around the z-axis.

Single-Image Depth Prediction Makes Feature Matching Easier

1 code implementation21 Aug 2020 Carl Toft, Daniyar Turmukhambetov, Torsten Sattler, Fredrik Kahl, Gabriel Brostow

Good local features improve the robustness of many 3D re-localization and multi-view reconstruction pipelines.

Depth Estimation Depth Prediction

Pose Proposal Critic: Robust Pose Refinement by Learning Reprojection Errors

no code implementations BMVC 2020 Lucas Brynte, Fredrik Kahl

In recent years, considerable progress has been made for the task of rigid object pose estimation from a single RGB-image, but achieving robustness to partial occlusions remains a challenging problem.

6D Pose Estimation using RGB

Fine-Grained Segmentation Networks: Self-Supervised Segmentation for Improved Long-Term Visual Localization

1 code implementation18 Aug 2019 Måns Larsson, Erik Stenborg, Carl Toft, Lars Hammarstrand, Torsten Sattler, Fredrik Kahl

In this paper, we propose a new neural network, the Fine-Grained Segmentation Network (FGSN), that can be used to provide image segmentations with a larger number of labels and can be trained in a self-supervised fashion.

Autonomous Driving Visual Localization

Monocular 3D Object Detection and Box Fitting Trained End-to-End Using Intersection-over-Union Loss

no code implementations19 Jun 2019 Eskil Jörgensen, Christopher Zach, Fredrik Kahl

We show how modeling heteroscedastic uncertainty improves performance upon our baseline, and furthermore, how back-propagation can be done through the optimizer in order to train the pipeline end-to-end for additional accuracy.

Autonomous Driving Monocular 3D Object Detection +1

A Cross-Season Correspondence Dataset for Robust Semantic Segmentation

1 code implementation16 Mar 2019 Måns Larsson, Erik Stenborg, Lars Hammarstrand, Torsten Sattler, Mark Pollefeys, Fredrik Kahl

We show that adding the correspondences as extra supervision during training improves the segmentation performance of the convolutional neural network, making it more robust to seasonal changes and weather conditions.

Semantic Segmentation

Semantic Match Consistency for Long-Term Visual Localization

no code implementations ECCV 2018 Carl Toft, Erik Stenborg, Lars Hammarstrand, Lucas Brynte, Marc Pollefeys, Torsten Sattler, Fredrik Kahl

Robust and accurate visual localization across large appearance variations due to changes in time of day, seasons, or changes of the environment is a challenging problem which is of importance to application areas such as navigation of autonomous robots.

Visual Localization

Rotation Averaging and Strong Duality

no code implementations CVPR 2018 Anders Eriksson, Carl Olsson, Fredrik Kahl, Tat-Jun Chin

In this paper we explore the role of duality principles within the problem of rotation averaging, a fundamental task in a wide range of computer vision applications.

A Projected Gradient Descent Method for CRF Inference allowing End-To-End Training of Arbitrary Pairwise Potentials

no code implementations24 Jan 2017 Måns Larsson, Anurag Arnab, Fredrik Kahl, Shuai Zheng, Philip Torr

It is empirically demonstrated that such learned potentials can improve segmentation accuracy and that certain label class interactions are indeed better modelled by a non-Gaussian potential.

Semantic Segmentation Structured Prediction

Globally Optimal Rigid Intensity Based Registration: A Fast Fourier Domain Approach

no code implementations CVPR 2016 Behrooz Nasihatkon, Frida Fejne, Fredrik Kahl

In this paper, we propose a dual algorithm in which the optimization is done in the Fourier domain, and multiple resolution levels are replaced by multiple frequency bands.

Minimizing the Maximal Rank

no code implementations CVPR 2016 Erik Bylow, Carl Olsson, Fredrik Kahl, Mikael Nilsson

In the latter case, matrices are divided into sub-matrices and the envelope is computed for each sub-block individually.


Optimal Relative Pose With Unknown Correspondences

no code implementations CVPR 2016 Johan Fredriksson, Viktor Larsson, Carl Olsson, Fredrik Kahl

Previous work on estimating the epipolar geometry of two views relies on being able to reliably match feature points based on appearance.

Multiresolution Search of the Rigid Motion Space for Intensity Based Registration

no code implementations14 Oct 2015 Behrooz Nasihatkon, Fredrik Kahl

Our results show that low resolution target values can tightly bound the high-resolution target function in natural images.

Image Registration

Minimal Solvers for Relative Pose with a Single Unknown Radial Distortion

no code implementations CVPR 2014 Yubin Kuang, Jan E. Solem, Fredrik Kahl, Kalle Astrom

In this paper, we study the problems of estimating relative pose between two cameras in the presence of radial distortion.

Accurate Localization and Pose Estimation for Large 3D Models

no code implementations CVPR 2014 Linus Svarm, Olof Enqvist, Magnus Oskarsson, Fredrik Kahl

For one, it makes the correspondence problem very difficult and it is likely that there will be a significant rate of outliers to handle.

Pose Estimation

Fast and Reliable Two-View Translation Estimation

no code implementations CVPR 2014 Johan Fredriksson, Olof Enqvist, Fredrik Kahl

It has long been recognized that one of the fundamental difficulties in theestimation of two-view epipolar geometry is the capability of handling outliers.

Motion Estimation Translation

Optimal Geometric Fitting under the Truncated L2-Norm

no code implementations CVPR 2013 Erik Ask, Olof Enqvist, Fredrik Kahl

First, it is shown that for a large class of problems, the statistically more desirable truncated L 2 -norm can be optimized with the same complexity.

Cannot find the paper you are looking for? You can Submit a new open access paper.