Robust Tracking Using Foreground-Background Texture Discrimination
International Journal of Computer Vision
Efficient adaptive density estimation per image pixel for the task of background subtraction
Pattern Recognition Letters
A model change detection approach to dynamic scene modeling
AVSS '09 Proceedings of the 2009 Sixth IEEE International Conference on Advanced Video and Signal Based Surveillance
Foundations and Trends in Robotics
Learning a scene background model via classification
IEEE Transactions on Signal Processing
Active curve axis Gaussian mixture models
Pattern Recognition
Cosaliency: where people look when comparing images
UIST '10 Proceedings of the 23nd annual ACM symposium on User interface software and technology
Background subtraction for automated multisensor surveillance: a comprehensive review
EURASIP Journal on Advances in Signal Processing - Special issue on advanced image processing for defense and security applications
Modeling and segmentation of floating foreground and background in videos
Pattern Recognition
Background updating for visual surveillance
ISVC'05 Proceedings of the First international conference on Advances in Visual Computing
Learning based system for detection and tracking of vehicles
SCIA'05 Proceedings of the 14th Scandinavian conference on Image Analysis
Intelligent multi-camera video surveillance: A review
Pattern Recognition Letters
Background subtraction using low rank and group sparsity constraints
ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part I
Moving object segmentation using motor signals
ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part V
ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part V
Efficient background subtraction for real-time tracking in embedded camera networks
Proceedings of the 10th ACM Conference on Embedded Network Sensor Systems
Spatiotemporal salience via centre-surround comparison of visual spacetime orientations
ACCV'12 Proceedings of the 11th Asian conference on Computer Vision - Volume Part III
A benchmark dataset for outdoor foreground/background extraction
ACCV'12 Proceedings of the 11th international conference on Computer Vision - Volume Part I
Hand segmentation for gesture recognition in EGO-vision
Proceedings of the 3rd ACM international workshop on Interactive multimedia on mobile & portable devices
| Hi-index | 0.00 |
Statistical background modelling and subtraction has proved to be apopular and effective class of algorithms for segmentingindependently moving foreground objects outfrom a staticbackground, without requiring any a priori information of theproperties of foreground objects. This paper presents twocontributions on this topic, aimed towards robotics where an activehead is mounted on a mobile vehicle. In periods when the vehicle'swheels are not driven, camera translation is virtually zero, andbackground subtraction techniques are applicable. Parts of thiswork are also highly relevant to surveillance and videoconferencing. The first part of the paper presents an efficientprobabilistic framework for when the camera pans and tilts. Aunified approach is developed for handling various sources oferror, including motion blur, sub-pixel camera motion, mixed pixelsat object boundaries, and also uncertainty in backgroundstabilisation caused by noise, unmodelled radial distortion andsmall translations of the camera. The second contribution regards aBayesian approach to specifically incorporate uncertaintyconcerning whether the background has yet been uncovered by movingforeground objects. This is an important requirement duringinitialisation of a system. We cannot assume that a backgroundmodel is available in advance since that would involve storingmodels for each possible position, in every room, of the robot'soperating environment. Instead the background model must begenerated online, very possibly in the presence of moving objects.