3 Advanced transfer learning use-cases for computer vision: part-i
This chapter covers
- A brief overview of advanced computer vision tasks such as search, object detection and segmentation
- Developing image search and de-duplication systems leveraging transfer learning concepts
- Building a high-level understanding of object detection and image segmentation tasks.
- Exploring hands-on examples using Transfer Learning in computer vision.
Computer vision is a domain where deep learning architectures have been battletested to outperform any of the previously known techniques. In the previous chapter, we developed an understanding of different computer vision tasks such as classification, search, object detection, segmentation and so on. We primarily focused upon classification task and discussed about a number of deep learning architectures. We also leveraged transfer learning to improve classification performance using the latest and the greatest pre-trained models. In this chapter, we will build upon the fundamentals of transfer learning and computer vision from the previous chapter and focus on some of the advanced computer vision tasks.