A Diffusion Mechanism for Obstacle Detection from Size-Change Information
IEEE Transactions on Pattern Analysis and Machine Intelligence
Image divergence and deformation from closed curves
International Journal of Robotics Research
Introductory Techniques for 3-D Computer Vision
Introductory Techniques for 3-D Computer Vision
Theory of Reconstruction from Image Motion
Theory of Reconstruction from Image Motion
What Is Computed by Structure from Motion Algorithms?
ECCV '98 Proceedings of the 5th European Conference on Computer Vision-Volume I - Volume I
A Qualitative Image Reconstruction from an Axial Image Sequence
AIPR '01 Proceedings of the 30th on Applied Imagery Pattern Recognition Workshop
Green Theorems and Qualitative Properties of the Optical Flow
Green Theorems and Qualitative Properties of the Optical Flow
| Hi-index | 0.00 |
This paper presents a global method toprocess monocular image sequences for mobile robotobstacle detection. We do not aim to achieve acomplete scene reconstruction, but only to evaluatethe time to collision and surface orientation useful forrobot obstacle avoidance.Using a planar facet representation we firstcalculate formally the velocity field generated by thecamera undergoing an arbitrary motion. Then, wepropose to interpret this vector flow along imagecontours, using an integrative approach inconjunction with a projective model, and weintroduce a link with image moments. Under theassumption of an approximate axial motion, we showthat our integrative approach can be applied as wellto estimate the image vector flow without involvingany prior feature correspondence. Noise resistance isimproved by integration. Experimental reconstructionresults on images sequences are given, and theirsensitivity to the various perturbations of cameramotion is evaluated.Potential applications include visual navigation,obstacle detection, visual servoing, depth fromzooming, object recognition.