No abstract

Abstract. We present a novel approach to segmenting video using iterated graph cuts based on spatio-temporal volumes. We use the mean shift clustering algorithm to build the spatio-temporal volumes with different bandwidths from the input video. We compute the prior probability obtained by the likelihood from a color histogram and a distance transform using the segmentation results from graph cuts in the previous process, and set the probability as the t-link of the graph for the next process. The proposed method can segment regions of an object with a stepwise process from global to local segmentation by iterating the graph-cuts process with mean shift clustering using a different bandwidth. It is possible to reduce the number of nodes and edges to about 1/25 compared to the conventional method with the same segmentation rate.

No abstract

It is important to make its vision system more robust and accurate, to give optimal visual-feedback, which helps to control a robot. We propose a robust and accurate pattern matching method for simultaneously identifying robots and estimating their orientations that does not use color segmentation. To search for similar patterns, our approach uses continuous DP matching, which is obtained by scanning at an ellipse circumference from the center of the robot. The DP similarity value is used to identify object, and to obtain the optimal route by back tracing to estimate its orientation. We found that our system's ability to identify objects was robust to variation in light conditions. This is because it can take advantage of the changes in intensity only. [4]

No abstract