Three-dimensional object recognition from single two-dimensional images
Artificial Intelligence
Fitting Parameterized Three-Dimensional Models to Images
IEEE Transactions on Pattern Analysis and Machine Intelligence
Empirical evaluation of dissimilarity measures for color and texture
Computer Vision and Image Understanding - Special issue on empirical evaluation of computer vision algorithms
Distinctive Image Features from Scale-Invariant Keypoints
International Journal of Computer Vision
Stable Real-Time 3D Tracking Using Online and Offline Information
IEEE Transactions on Pattern Analysis and Machine Intelligence
Improved Adaptive Gaussian Mixture Model for Background Subtraction
ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 2 - Volume 02
Application of Deterministic Low-Discrepancy Sequences in Global Optimization
Computational Optimization and Applications
A Comparison of Affine Region Detectors
International Journal of Computer Vision
International Journal of Computer Vision
Photo tourism: exploring photo collections in 3D
ACM SIGGRAPH 2006 Papers
Web-based 3D Reconstruction Service
Machine Vision and Applications
Object recognition and full pose registration from a single image for robotic manipulation
ICRA'09 Proceedings of the 2009 IEEE international conference on Robotics and Automation
Location recognition using prioritized feature matching
ECCV'10 Proceedings of the 11th European conference on Computer vision: Part II
CVPR '11 Proceedings of the 2011 IEEE Conference on Computer Vision and Pattern Recognition
Hi-index | 0.00 |
Determining the pose of objects appearing in images is a problem encountered often in several practical applications. The most effective strategy for dealing with this challenge is to proceed according to the model-based paradigm, which involves building 3D models of objects and then determining object poses by fitting their models to new images with the aid of detected features. This paper proposes a model-based approach for estimating the full pose of known objects from natural point features. The method employs a projective imaging model and incorporates reliable automatic mechanisms for pose initialization and convergence. Furthermore, it is extendable to multiple cameras without the need to perform multi-view matching and relies on sparse structure from motion techniques for the construction of object models offline. Experimental results demonstrate its accuracy and robustness.