Camera auto-calibration using a sequence of 2D images with small rotations and translations

Date

2003-01-01

Author

Hassanpour, R
Atalay, Mehmet Volkan

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

8
views

0
downloads

3D model generation needs depth information of the object in the input images. This information can be found using stereo imaging but it needs camera parameters. Camera calibration is not possible without some knowledge about the objects in the scene or assuming fixed or known values for the camera parameters. When using fixed camera parameters, however, small rotation angles or small translation in camera position can degenerate the results. The degeneracy can be omitted by adding new restrictions to the a-priori knowledge about the camera parameters. The calibrated data may be used to reconstruct 3D model of the scene.

URI

https://hdl.handle.net/11511/54168

Collections

Department of Computer Engineering, Conference / Seminar

Suggestions

OpenMETU
Core

Experimental study on the sensitivity of autocalibration to projective camera model parameters Hassanpour, Reza; Atalay, Mehmet Volkan (2006-04-01) Existing methods of 3-D object modeling and recovering 3-D data from uncalibrated 2-D images are subject to errors introduced by assumptions about camera parameters and mismatches in finding point pairs in the images. In this study, we experimentally evaluate the effect of each of these assumptions together with the inaccuracy in the measurements in the images. Sensitivity of reconstruction errors to inaccuracies in the estimation of camera parameters and mismatches due to noise in input data is measured us...
3D shape recovery and tracking from multi-camera video sequences via surface deformation Çok kamerali video görüntülerinden yüzey deformasyonu ile 3B şekil geric̈atma ve i̇zleme Sahillioğlu, Yusuf; Yemez, Y.; Skala, V. (2006-12-01) This paper addresses 3D reconstruction and modeling of time-varying real objects using multicamera video. The work consists of two phases. In the first phase, the initial shape of the object is recovered from its silhouettes using a surface deformation model. The same deformation model is also employed in the second phase to track the recovered initial shape through the time-varying silhouette information by surface evolution. The surface deformation/evolution model allows us to construct a spatially and te...
Extraction of 3D transform and scale invariant patches from range scans Akagunduz, Erdern; Ulusoy, İlkay (2007-06-22) An algorithm is proposed to extract transformation and scale invariant 3D fundamental elements from the surface structure of 3D range scan data. The surface is described by mean and Gaussian curvature values at every data point at various scales and a scale-space search is performed in order to extract the fundamental structures and to estimate the location and the scale of each fundamental structure. The extracted fundamental structures can later be used as nodes in a topological graph where the links betw...
MARS - A TOOL-BASED MODELING, ANIMATION, AND PARALLEL RENDERING SYSTEM AKTIHANOGLU, M; OZGUC, B; AYKANAT, C (Springer Science and Business Media LLC, 1994-01-01) This paper describes a system for modeling, animating, previewing and rendering articulated objects. The system has a modeler of objects that consists of joints and segments. The animator interactively positions the articulated object in its stick, control vertex, or rectangular prism representation and previews the motion in real time. Then the data representing the motion and the models is sent to a multicomputer [iPSC/2 Hypercube (Intel)]. The frames are rendered in parallel, exploiting the coherence bet...
Visual object tracking using semi supervised convolutional filters Sevindik, Emir Can; Alatan, Abdullah Aydın; Department of Electrical and Electronics Engineering (2020-10-15) Visual object tracking aims to find a single object position in a video frame, when a annotated bounding box is provided in the first frame. Correlation filters have always produced excellent results in terms of accuracy, while enjoying quite low computational complexity. The main property of correlation filter based trackers is to find a filter that can generate high values around the true target object location, whereas relatively low values for the locations away from the object. Recently, deep learn...

Citation Formats

R. Hassanpour and M. V. Atalay, “Camera auto-calibration using a sequence of 2D images with small rotations and translations,” 2003, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/54168.