Vanishing point calculation as a statistical inference on the unit sphere

Collins, Robert T.; Weiss, Richard

doi:10.1109/iccv.1990.139560

Cited by 109 publications

(66 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…3 Any vanishing point (x, y, 1) on the horizon line associated with the ground plane (and every other horizontal plane in the scene) is orthogonal with v 0 , which gives the constraint [11] …”

Section: Camera Geometrymentioning

confidence: 99%

See 1 more Smart Citation

Automatic Surveillance Camera Calibration without Pedestrian Tracking

Liu¹,

Collins²,

Liu³

2011

Procedings of the British Machine Vision Conference 2011

View full text Add to dashboard Cite

We propose a new framework for automatic surveillance camera calibration by observing videos of pedestrians walking through the scene. Unlike existing methods that require accurate pedestrian detection and tracking, our method takes noisy foreground masks as input and automatically estimates the necessary intrinsic and extrinsic camera parameters using prior knowledge about the distribution of relative human heights. Our algorithm is computationally efficient enough for online parameter estimation. Experimental results on both synthetic and real data show the robustness of our method to camera pose and noisy foreground detections.

show abstract

Section: Camera Geometrymentioning

confidence: 99%

“…Following RANSAC selection, we use the method of [2,3] to fit a least squares vanishing point estimate to the inlier blobs. The major axes of all K inlier ellipses are organized into a matrix with normalized homogeneous coordinate representation:…”

Section: Vanishing Point Estimation Via Foreground Blobsmentioning

confidence: 99%

Automatic Surveillance Camera Calibration without Pedestrian Tracking

Liu¹,

Collins²,

Liu³

2011

Procedings of the British Machine Vision Conference 2011

View full text Add to dashboard Cite

show abstract

“…When the camera is calibrated, the image line segments are represented as unit vectors on the Gaussian sphere and several techniques for both grouping and initialization stage on the Gaussian sphere exist [1][2][3]. The main advantage of the Gaussian sphere representation is the equal treatment of all possible vanishing directions, including those at infinity.…”

Section: Vanishing Point Estimationmentioning

confidence: 99%

“…Previous techniques for line segment grouping vary in the choice of the accumulator space, where the peaks correspond to the dominant clusters of line segments; most common alternatives are the Gaussian sphere and Hough space [1,3,5,7]. When the camera is calibrated, the image line segments are represented as unit vectors on the Gaussian sphere and several techniques for both grouping and initialization stage on the Gaussian sphere exist [1][2][3].…”

Section: Vanishing Point Estimationmentioning

confidence: 99%

Extraction, matching and pose recovery based on dominant rectangular structures

Zhang

Košecká

2003

First IEEE International Workshop on Higher-Level Knowledge in 3D Modeling and Motion Analysis, 2003. HLK 2003.

View full text Add to dashboard Cite

Man-made environments possess many regularities which can be efficiently exploited for image based rendering as well as robotic navigation and localization tasks. In this paper we present an approach for automatic extraction of dominant rectangular structures from a single view and show how they facilitate the recovery of camera pose, planar structure and matching across widely separated views. In the presented approach the rectangular hypothesis formation is based on a higher level information encoded by the presence of orthogonal vanishing directions, the dominant rectangular structures can be detected and matched despite the presence of multiple repetitive structures often encountered in a variety of buildings. Different stages of the approach are demonstrated on various examples of images of indoors and outdoors structured environments.

show abstract

“…Since lines assumed parallel in the 3D space have their images only approximately concurrent in the image plane, techniques have been developed to model the orientation error [6], as well as fully statistical approaches for managing uncertainty (such as those described in [7] and [8]). Improvements can also be obtained when the geometry of objects of interest in the scene is known, by exploiting primitive-based techniques [6].…”

Section: Gaussian-sphere-based Approachesmentioning

confidence: 99%