A Framework for High-Level Feedback to Adaptive, Per-Pixel, Mixture-of-Gaussian Background Models
ECCV '02 Proceedings of the 7th European Conference on Computer Vision-Part III
A Framework for Robust Subspace Learning
International Journal of Computer Vision - Special Issue on Computational Vision at Brown University
A Hierarchical Approach to Robust Background Subtraction using Color and Gradient Information
MOTION '02 Proceedings of the Workshop on Motion and Video Computing
Integrated Region- and Pixel-based Approach to Background Modelling
MOTION '02 Proceedings of the Workshop on Motion and Video Computing
Robust Face Recognition via Sparse Representation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Motion-based background subtraction using adaptive kernel density estimation
CVPR'04 Proceedings of the 2004 IEEE computer society conference on Computer vision and pattern recognition
Decoding by linear programming
IEEE Transactions on Information Theory
Near-Optimal Signal Recovery From Random Projections: Universal Encoding Strategies?
IEEE Transactions on Information Theory
Statistical modeling of complex backgrounds for foreground object detection
IEEE Transactions on Image Processing
Background subtraction based on phase feature and distance transform
Pattern Recognition Letters
Hi-index | 0.00 |
We investigate effective means of building robust dictionaries for detecting the sparse foreground in videos with static background. This work is an extension to our existing solution [1] to foreground/background segmentation problem using the linear programming method [2] proposed to detect sparse errors in signals, which are created by a known dictionary. The dictionary building methods we study are established robust component analysis techniques in the literature (i.e. k-SVD [3] & robust-PCA [4]) as well as a heuristic (running median) inspired by the highly correlated nature of the static video background signal. We compare the effectiveness of the new methods with our original system as well as a baseline method, which is the commonly used single Gaussian model of the background pixels.