Visual Object Tracking using Deep Learning

The text comprehensively discusses tracking architecture under stochastic and deterministic frameworks and presents experimental results under each framework with qualitative and quantitative analysis. It covers deep learning techniques for feature extraction, template matching, and training the networks in tracking algorithms. Discusses performance metrics for visual tracking in comparing the efficiency and effectiveness of available datasets. Covers performance metrics such as center location error, F-Measure, area under control, distance precision, and overlap precision. Compares the performance of deep learning trackers with traditional methods, wherein hand-crafted features were fused to reduce the computational complexity. Illustrates stochastic framework for visual tracking such as probabilistic methods in the Bayesian framework for state estimation. The text presents both traditional and advanced methods such as stochastic, deterministic, generative, discriminative framework, and deep learning-based appearance models. It further highlights the use of deep learning for feature extraction, template matching, and training the networks in tracking algorithms. The book covers context-aware, and super pixel-based correlation filter tracking. The text is primarily written for senior undergraduate, graduate students, and academic researchers in the fields of electrical engineering, electronics and communication engineering, computer engineering, and information technology.