Boosted Ringlet Features for Visual Object Tracking
Almabrok Essa Essa, Evan W Krieger
Accurate and efficient object tracking is an important aspect in security and surveillance applications. Many challenges exist in visual object tracking including scale change, object distortion, lighting change, and occlusion. The combination of structural target information including edge features with the intensity or color features will allow for more robust object tracking in these conditions. To achieve this, we propose a feature extraction method that utilizes both the Frei-Chen edge detector and Gaussian ringlet feature mapping. Frei-Chen edge detector extracts edge, line, and mean features that can be combined to create an edge detection image. The edge detection image will then be used to represent the structural features of the target. Gaussian ringlet feature mapping is used to obtain rotational invariant features that are robust to target and viewpoint rotation. These aspects are combined to create an efficient and robust tracking scheme. The proposed method also includes occlusion and scale handling components. The proposed scheme is evaluated against state-of-the-art feature tracking methods using both temporal and spatial robustness metrics on the Visual Object Tracking 2014 database.
Graduate Research - Graduate
Vijayan K Asari
Primary Advisor's Department
Electrical and Computer Engineering
Stander Symposium poster
"Boosted Ringlet Features for Visual Object Tracking" (2017). Stander Symposium Posters. 1024.