Minimal Solvers for Single-View Lens-Distorted Camera Auto-Calibration

Lochman, Yaroslava; Dobosevych, Oles; Hryniv, Rostyslav; Pritts, James

doi:10.1109/wacv48630.2021.00293

Cited by 7 publications

(4 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most sites are observed with two web cameras only and thus there are two possible approaches for performing the autocalibration. One is using a single image and makes some assumptions about the scene content [21,16], e.g. looks for repetitive patterns or a Manhattan frame, the other considers two views of a scene and computes the fundamental matrix along with the distortion parameters [13].…”

Section: Camera Calibrationmentioning

confidence: 99%

“…For the single-view approach, the hybrid solver with six circular arcs (Solver 6CA) proposed by [16] is adopted, which takes an input image and recovers the parameters by identifying Manhattan Frames (MF) in the image. The algorithm deals with two configurations in particular: 1) the vanishing points are coplanar; 2) the vanishing points are mutually orthogonal.…”

Section: Camera Calibrationmentioning

confidence: 99%

“…Fortunately, many state-of-the-art approaches were developed to handle the above challenges. For example, efficient and robust calibration of camera intrinsics including radial distortion became recently available from two [13,14] or even a single image [21,16]. Additionally, the image matching challenge posed by large baselines for outdoor environments is now a popular research topic and many deep learning-based approaches have boosted the matching performances [22] even in extreme matching conditions [12].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

3D Reconstruction from public webcams

Wu¹,

Schindler²,

Albl³

2021

Preprint

View full text Add to dashboard Cite

In this paper, we investigate the possibility of reconstructing the 3D geometry of a scene captured by multiple webcams. The number of publicly accessible webcams is already large and it is growing every day. A logical question arises -can we use this free source of data for something beyond leisure activities? The challenge is that no internal, external, or temporal calibration of these cameras is available. We show that using recent advances in computer vision, we successfully calibrate the cameras, perform 3D reconstructions of the static scene and also recover the 3D trajectories of moving objects.

show abstract

Section: Camera Calibrationmentioning

confidence: 99%

Section: Camera Calibrationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

3D Reconstruction from public webcams

Wu¹,

Schindler²,

Albl³

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The use of dedicated images ("training data") for the task of camera calibration distinguishes standard calibration from selfcalibration, that extracts calibration parameters from uncontrolled "test" images (e.g. [9,17,30,12,42,31,24]).…”

Section: Related Workmentioning

confidence: 99%

BabelCalib: A Universal Approach to Calibrating Central Cameras

Lochman

Liepieshov

Chen

et al. 2021

2021 IEEE/CVF International Conference on Computer Vision (ICCV)

View full text Add to dashboard Cite

Existing calibration methods occasionally fail for large field-of-view cameras due to the non-linearity of the underlying problem and the lack of good initial values for all parameters of the used camera model. This might occur because a simpler projection model is assumed in an initial step, or a poor initial guess for the internal parameters is pre-defined. A lot of the difficulties of general camera calibration lie in the use of a forward projection model. We side-step these challenges by first proposing a solver to calibrate the parameters in terms of a back-projection model and then regress the parameters for a target forward model. These steps are incorporated in a robust estimation framework to cope with outlying detections. Extensive experiments demonstrate that our approach is very reliable and returns the most accurate calibration parameters as measured on the downstream task of absolute pose estimation on test sets. The code is released at https://github.com/ylochman/babelcalib.

show abstract