All interested are welcome to attend the below seminar:
Time/Date: Friday 10th August 2012, at 3.05pm.
Venue: B21, Teaching Suite 5 – Ingarkni Wardli Building
Title: Robust 3D optic flow estimation with direct smoothing control
Abstract: Complexities of dynamic volumetric imaging challenge the available computer vision techniques on a number of different fronts. In this talk, the relationship between the estimation accuracy and required amount of smoothness for a general optic flow solution will be explored. Our experimental results have shown that surprisingly a small amount of local smoothing is required to satisfy both the necessary and sufficient conditions for accurate optic flow estimation. This notion is called “just enough” smoothing and its proper implementation is expected to have a significant effect on the preservation of local information in processing 3D dynamic scans. To demonstrate the effect of “just enough” smoothing, a robust 3D optic flow method with direct smoothing control was devised. The effect of local smoothing on the accuracy of motion estimation in dynamic lung CT images will be highlighted using the results of this method on both synthetic and real image sequences with ground truth.
Biographical details: Ali Bab-Hadiashar is currently a professor of Mechatronics at RMIT University, Melbourne. He has been involved in international Mechatronics engineering education for over a decade. He was a founding participant and a member of the technical and executive committees of the Research Centre for Advanced By-Wire Technologies (RABiT) which received over $5M of government funding from 2003-2007.
His research interests include statistical segmentation of visual data, use of unmanned aerial vehicles for environmental monitoring, intelligent and robust sensory systems for new generation vehicle technologies and robot systems for advanced manufacturing. He has attracted a number of research grants (including ARC SPIRT, Linkage and Discovery) and has published several highly cited papers in optic flow, robust segmentation, model selection and data fusion techniques. He is a senior member of IEEE and has been part of organising and program committees of several national and international conferences including the IEEE Conference on Image Processing (ICIP2013) to be held in Melbourne in 2013.