Clustering in video data: Dealing with heterogeneous semantics of features
Pattern Recognition
Local Adaptivity to Variable Smoothness for Exemplar-Based Image Regularization and Representation
International Journal of Computer Vision
Track and cut: simultaneous tracking and segmentation of multiple objects with graph cuts
Journal on Image and Video Processing - Video Tracking in Complex Scenes for Surveillance Applications
Robust NL-means filter with optimal pixel-wise smoothing parameter for statistical image denoising
IEEE Transactions on Signal Processing
Improved object tracking using an adaptive colour model
EMMCVPR'07 Proceedings of the 6th international conference on Energy minimization methods in computer vision and pattern recognition
A framework for unsupervised segmentation of multi-modal medical images
CVAMIA'06 Proceedings of the Second ECCV international conference on Computer Vision Approaches to Medical Image Analysis
| Hi-index | 0.00 |
We consider the problem of segmentation of images that can bemodelled as piecewise continuous signals having unknown,non-stationary statistics. We propose a solution to this problemwhich first uses a regression framework to estimate the image PDF,and then mean-shift to find the modes of this PDF. The segmentationfollows from mode identification wherein pixel clusters or imagesegments are identified with unique modes of the multi-modal PDF.Each pixel is mapped to a mode using a convergent, iterativeprocess. The effectiveness of the approach depends upon theaccuracy of the (implicit) estimate of the underlying multi-modaldensity function and thus on the bandwidth parameters used for itsestimate using Parzen windows. Automatic selection of bandwidthparameters is a desired feature of the algorithm. We show that theproposed regression-based model admits a realistic framework toautomatically choose bandwidth parameters which minimizes a globalerror criterion. We validate the theory presented with results onreal images.