Robust vanishing point determination in noisy images

Bräuer-Burchardt, Christian; Voss, Klaus

doi:10.1109/icpr.2000.905399

Cited by 21 publications

(15 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To obtain the correct axis inclination the vertical vanishing point (computed by a robust technique as described in [50]) is then used as shown in Fig. 1.…”

Section: Multiple Camera Processingmentioning

confidence: 99%

Detecting anomalies in people’s trajectories using spectral graph analysis

Calderara

Heinemann

Prati

et al. 2011

Computer Vision and Image Understanding

View full text Add to dashboard Cite

“…To obtain the correct axis inclination the vertical vanishing point (computed by a robust technique as described in [50]) is then used as shown in Fig. 1.…”

Section: Multiple Camera Processingmentioning

confidence: 99%

Detecting anomalies in people’s trajectories using spectral graph analysis

Calderara

Heinemann

Prati

et al. 2011

Computer Vision and Image Understanding

View full text Add to dashboard Cite

“…This is not true for general setups, but the correct axis orientation can be obtained by computing the position of the camera's vertical vanishing point vp. The vanishing points are computed with the procedure described in [1]. In this way, the warped axis is obtained as the segment between the warped lower support point lp and the intersection with the epipolar line.…”

Section: Likelihood Computationmentioning

confidence: 99%

Group Detection at Camera Handoff for Collecting People Appearance in Multi-camera Systems

Calderara

Cucchiara

Prati

2006

2006 IEEE International Conference on Video and Signal Based Surveillance

View full text Add to dashboard Cite

Logging information on moving objects is crucial in video surveillance systems. Distributed multi-camera systems can provide the appearance of objects/people from different viewpoints and at different resolutions, allowing a more complete and precise logging of the information. This is achieved through consistent labeling to correlate collected information of the same person. This paper proposes a novel approach to consistent labeling also capable to fully characterize groups of people and to manage miss segmentations. The ground-plane homography and the epipolar geometry are automatically learned and exploited to warp objects' principal axes between overlapped cameras. A MAP estimator that exploits two contributions (forward and backward) is used to choose the most probable label configuration to be assigned at the handoff of a new object. Extensive experiments demonstrate the accuracy of the proposed method in detecting single and simultaneous handoffs, miss segmentations, and groups.

show abstract

“…Camera calibration has been studied for many years and there are many methods available to find the parameters precisely. Although most existing methods require information of the known scene points and the use of multiple images, there are several calibration approaches exploiting the presence of vanishing points which have been reported in close‐range photogrammetry (12,25,26) as well as in computer vision (16,27–29). In the context of close‐range photogrammetry the general approach established is based on the use of three orthogonal vanishing points and some constraints among lines, while in the context of computer vision, several approaches are supported by the computation and decomposition of the absolute conic from three vanishing points or the rotation matrix.…”

Section: Photogrammetric Processmentioning

confidence: 99%

Forensic Terrestrial Photogrammetry from a Single Image

Aguilera

Gómez-Lahoz

2009

Journal of Forensic Sciences

View full text Add to dashboard Cite

Forensic terrestrial photogrammetry is one of the most valuable and low-cost resources of spatial data available today. Due to the ephemeral crime scene characteristics, these photographs can often capture information that is never to be seen again. This paper presents a novelty approach for the documentation, analysis, and visualization of crime scenes for which only a single perspective image is available. The photogrammetric process consists of a few well-known steps in close-range photogrammetry: features extraction, vanishing points computation, camera self-calibration, 3D metric reconstruction, dimensional analysis, and interactive visualization. Likewise, the method incorporates a quality control of the different steps accomplished sequentially. As a result, several cases of study are presented in the experimental results section in order to test their viability. The full approach can be applied easily through the free software, sv3DVision, which has been evaluated by a number of police officers, forensic scientists, and forensic educators satisfactorily.

show abstract

Robust vanishing point determination in noisy images

Cited by 21 publications

References 9 publications

Detecting anomalies in people’s trajectories using spectral graph analysis

Detecting anomalies in people’s trajectories using spectral graph analysis

Group Detection at Camera Handoff for Collecting People Appearance in Multi-camera Systems

Forensic Terrestrial Photogrammetry from a Single Image

Contact Info

Product

Resources

About