Statistics of Flow Vectors and Its Application to the Voting Method for the Detection of Flow Fields

Authors:
Atsushi Imiya;Keisuke Iwawaki
Affiliations:
-;-
Venue:
MLDM '01 Proceedings of the Second International Workshop on Machine Learning and Data Mining in Pattern Recognition
Year:
2001

Citing 5
Cited 0

The computation of optical flow

ACM Computing Surveys (CSUR)
Recovering 3-D motion parameters from optical flow field using randomized Hough transform

Pattern Recognition Letters
The medians of discrete sets

Information Processing Letters
Motion analysis by random sampling and voting process

Computer Vision and Image Understanding
Statistical Optimization and Geometric Visual Inference

AFPAC '97 Proceedings of the International Workshop on Algebraic Frames for the Perception-Action Cycle

Quantified Score

Hi-index	0.01

Visualization

Abstract

In this paper, we show that the randomized sampling and voting process detects linear flow filed as a model-fitting problem. We introduce a random sampling method for solving the least-square model-fitting-problem using a mathematical property for the construction of pseudo-inverse. If we use an appropriate number of images from a sequence of images, it is possible to detect subpixel motion in this sequence. We use the accumulator space for the unification of these flow vectors which are computed from different time intervals. Numerical examples for the test image sequences show the performance of our method.