• P Jayanthi
• Shaik Taj Mahaboob
• Varna Kumar

Visual Tracking means locating a moving object or focusing on a particular object over time by using camera. It is a time-consuming process due to the amount of data contained in that video. So, we use object recognition techniques for tracking to reduce the complexity. In this paper, we have an advantage that we consider interactions of grids not only from same spatial positions, but also from different positions by using similarity learning. Target appearance with a fixed spatial grid layout without considering interaction between different grids, may affect their performance when the target appearance suffers from heavy occlusions, large deformation variations to overcome this we use nonlocal similarity learning. We use many applications for Visual Tracking such as Human computer interfaces, surveillance, traffic control, motion analysis and driverless vehicle.

Nonlocal similarity learning, Visual Tracking, particle filters, logistic regression.

"A COMPREHENSIVE STUDY OF VISUAL TRACKING VIA NONLOCAL SIMILARITY LEARNING", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 3, page no.298-300, March-2019, Available :http://www.jetir.org/papers/JETIRAK06056.pdf

"A COMPREHENSIVE STUDY OF VISUAL TRACKING VIA NONLOCAL SIMILARITY LEARNING", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 3, page no. pp298-300, March-2019, Available at : http://www.jetir.org/papers/JETIRAK06056.pdf