Abstract
In this chapter, we will look at some of the advanced concepts and models in deep learning. Image segmentation and object localization and detection are some of the key areas that have garnered a lot of importance lately. Image segmentation plays a crucial role in detecting diseases and abnormalities through the processing of medical images. At the same time, it is equally crucial in industries such as aviation, manufacturing, and other domains to detect anomalies such as cracks or other unwanted conditions in machinery. Alternately, images of the night sky can be segmented to detect previously unknown galaxies, stars, and planets. Object detection and localization has profound use in places requiring constant automated monitoring of activities, such as in shopping malls, local stores, industrial plants, and so on. Also, it can be used to count objects and people in an area of interest and estimate various densities, such as traffic conditions at various signals. We will begin this chapter by going through a few of the traditional methods of image segmentation so that we can appreciate how neural networks are different from their traditional counterparts. Then, we will look at object-detection and localization techniques, followed by generative adversarial networks, which have gained lot of popularity recently because of their use and potential as a generative model to create synthetic data. This synthetic data can be used for training and inference in case there is not much data available or the data is expensive to procure. Alternatively, generative models can be used for style transfer from one domain to another.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to APress Media, LLC, part of Springer Nature
About this chapter
Cite this chapter
Pattanayak, S. (2023). Advanced Neural Networks. In: Pro Deep Learning with TensorFlow 2.0. Apress, Berkeley, CA. https://doi.org/10.1007/978-1-4842-8931-0_6
Download citation
DOI: https://doi.org/10.1007/978-1-4842-8931-0_6
Published:
Publisher Name: Apress, Berkeley, CA
Print ISBN: 978-1-4842-8930-3
Online ISBN: 978-1-4842-8931-0
eBook Packages: Professional and Applied ComputingApress Access BooksProfessional and Applied Computing (R0)