Local feature matching is a critical part of many computer vision pipelines, including among others Structure-from-Motion, SLAM, and Visual Localization. However, due to limitations in the descriptors, raw matches are often contaminated by a majority of outliers. As a result, outlier detection is a fundamental problem in computer vision and a wide range of approaches, from simple checks based on d

In this paper we consider the problem of Structure-from-Motion from images with unknown intrinsic calibration. Instead of estimating the internal camera parameters through some self-calibration procedure, we propose to use a subset of the reprojection constraints that is invariant to radial displacement. This allows us to recover metric 3D reconstructions without explicitly estimating the cameras’

Camera pose estimation in known scenes is a 3D geometry task recently tackled by multiple learning algorithms. Many regress precise geometric quantities, like poses or 3D points, from an input image. This either fails to generalize to new viewpoints or ties the model parameters to a specific scene. In this paper, we go Back to the Feature: we argue that deep networks should focus on learning robus

Motion blur is one of the major challenges remaining for visual odometry methods. In low-light conditions where longer exposure times are necessary, motion blur can appear even for relatively slow camera motions. In this paper we present a novel hybrid visual odometry pipeline with direct approach that explicitly models and estimates the camera’s local trajectory within the exposure time. This all

Compared to feature point detection and description, detecting and matching line segments offer additional challenges. Yet, line features represent a promising complement to points for multi-view tasks. Lines are indeed well-defined by the image gradient, frequently appear even in poorly textured areas and offer robust structural cues. We thus hereby introduce the first joint detection and descrip

BACKGROUND CONTEXT: Degenerative cervical spondylotic myelopathy (DCM) is the commonest form of spinal cord injury in adults. However, a limited number of clinical reports have assessed the role of biomarkers in DCM. PURPOSE: We evaluated cerebrospinal fluid (CSF) biomarkers in patients scheduled for DCM surgery and hypothesized that CSF biomarkers levels (1) would reflect the severity of preopera

Recent works on localization and mapping from privacy preserving line features have made significant progress towards addressing the privacy concerns arising from cloud-based solutions in mixed reality and robotics. The requirement for calibrated cameras is a fundamental limitation for these approaches, which prevents their application in many crowd-sourced mapping scenarios. In this paper, we pro

Finding local features that are repeatable across multiple views is a cornerstone of sparse 3D reconstruction. The classical image matching paradigm detects keypoints per-image once and for all, which can yield poorly-localized features and propagate large errors to the final geometry. In this paper, we refine two key steps of structure-from-motion by a direct alignment of low-level image informat

Oceanic islands have isolated biota, which typically include many endemic species. However, island endemics are vulnerable due to small population sizes, and they are often threatened by habitat destruction or by introduced pests and predators. Adequate conservation planning requires good information on genetic variability and population structure, also when seemingly viable species are considered

We consider the problem of stitching image sequences with cameras undergoing pure rotational motion. We leverage the assumption of a locally constant rotation axis, i.e., neighboring frames have a shared but unknown rotation axis. This assumption holds in many common image capturing scenarios, e.g., panoramic sweeping motions. Using this additional constraint, we develop techniques for three-view

This paper presents a two-stage procedure to create a surface integrity predictor. The first stage includes data clustering, which allows to evaluate the achievable surface quality. The second stage consists in training the model to predict which cluster the machined surface will belong to. To demonstrate the applicability, an experimental plan for machining of Inconel 718 in milling was developed

Neural implicit representations have recently shown encouraging results in various domains, including promising progress in simultaneous localization and mapping (SLAM). Nevertheless, existing methods produce over- smoothed scene reconstructions and have difficulty scaling up to large scenes. These limitations are mainly due to their simple fully-connected network architecture that does not incorp

Low-dimensional parametric models are the de-facto standard in computer vision for intrinsic camera calibration. These models explicitly describe the mapping between incoming viewing rays and image pixels. In this paper, we explore an alternative approach which implicitly models the lens distortion. The main idea is to replace the parametric model with a regularization term that ensures the latent

Translation of the poem "Dancing in Odessa" by Ukrainian-American poet Ilya Kaminsky (b. 1977)Translation of the poem "Dancing in Odessa" by Ukrainian-American poet Ilya Kaminsky (b. 1977)

Visibility across supply chains has been a key concern for organisations for many years, but the tools and information systems to achieve real-time visibility have not been available until recently. In response to uncertainty and complexity, advanced information and communication technologies have been explored for supply chain visibility (SCV). However, managerial perspectives are largely absent

The Veiki moraine in northern Sweden, a geomorphologically distinct landscape of ice-walled lake plains, has been interpreted to represent the former margin of an intermediate-sized pre–Last Glacial Maximum (LGM) Fennoscandian ice sheet, but its age is debated as either marine isotope stage (MIS) 5c or MIS 3. We have applied optically stimulated luminescence (OSL) and radiocarbon dating to four si

Contact-rich manipulation tasks remain a hard problem in robotics that requires interaction with unstructured environments. Reinforcement Learning (RL) is one potential solution to such problems, as it has been successfully demonstrated on complex continuous control tasks. Nevertheless, current state-of-the-art methods require policy training in simulation to prevent undesired behavior and later d

AimTo investigate the complete clinical spectrum of individuals with paediatric tuberous sclerosis complex in southern Sweden and explore changes over time.MethodsIn this retrospective observational study, 52 individuals aged up to 18 years at the study start were followed-up at regional hospitals and centres for habilitation from 2000 to 2020.ResultsCardiac rhabdomyoma was detected prenatally/neo