|Название: Visual Object Tracking using Deep Learning
Автор: Ashish Kumar
Издательство: CRC Press
Формат: pdf (true)
Размер: 20.3 MB
This book covers the description of both conventional methods and advanced methods. In conventional methods, visual tracking techniques such as stochastic, deterministic, generative, and discriminative are discussed. The conventional techniques are further explored for multi-stage and collaborative frameworks. In advanced methods, various categories of deep learning-based trackers and correlation filter-based trackers are analyzed.
Over the past few years, visual tracking algorithms are evolved from conventional frameworks to deep learning-based frameworks. Conventional trackers are fast in processing but are not effective for tedious environmental variations. Conventional trackers are explored as probabilistic, generative, discriminative, and deterministic frameworks. In a direction to improve tracking accuracy, the collaborative and multi-stage tracking frameworks are analyzed for real-time outcomes for tedious tracking problems.
Deep Learning trackers provide discriminative features with efficient performance in complex tracking variations. It became necessary to extract the salient features in target appearance models using various deep neural networks. Handcrafted features such as color, texture, and HOG are integrated with deep features to improve their processing speed so that solutions to real-time tracking problems can be provided. Hyper-features are also extracted from various layers of deep neural network to address a wide range of tracking solutions.
The book also:
- Discusses potential performance metrics used for comparing the efficiency and effectiveness of various visual tracking methods.
- Elaborates on the salient features of deep learning trackers along with traditional trackers, wherein the handcrafted features are fused to reduce computational complexity.
- Illustrates various categories of correlation filter-based trackers suitable for superior and efficient performance under tedious tracking scenarios.
- Explores the future research directions for visual tracking by analyzing the real-time applications.
The book comprehensively discusses various deep learning-based tracking architectures along with conventional tracking methods. It covers in-depth analysis of various feature extraction techniques, evaluation metrics and benchmark available for performance evaluation of tracking frameworks. The text is primarily written for senior undergraduates, graduate students, and academic researchers in the fields of electrical engineering, electronics and communication engineering, computer engineering, and information technology.
Скачать Visual Object Tracking using Deep Learning