Fine-grained recognition : data, recognition, and application
Abstract/Contents
- Abstract
- Fine-grained recognition refers to the task in computer vision of automatically differentiating similar object categories from one another, e.g. species of birds, types of cars, breeds of dogs, or varieties of aircraft. Applications are numerous: beyond simply being able to describe the world in more detail, fine-grained recognition can be used for improved scene understanding, studying society, and even analyzing biodiversity. Unfortunately, fine-grained recognition is extremely challenging. Even acquiring data is difficult due to the expertise required in annotation, and differences between categories can still be extremely subtle and formidable to learn. In this work, I will describe the efforts I have made toward tackling fine-grained recognition. These include four projects aimed at improving automatic recognition of fine-grained categories, which look at both the data and recognition algorithms used. I also describe an application of fine-grained recognition in understanding society from images. Finally, I present a work looking at finer-grained natural language descriptions of images.
Description
Type of resource | text |
---|---|
Form | electronic; electronic resource; remote |
Extent | 1 online resource. |
Publication date | 2016 |
Issuance | monographic |
Language | English |
Creators/Contributors
Associated with | Krause, Jonathan |
---|---|
Associated with | Stanford University, Department of Computer Science. |
Primary advisor | Li, Fei Fei, 1976- |
Thesis advisor | Li, Fei Fei, 1976- |
Thesis advisor | Ermon, Stefano |
Thesis advisor | Savarese, Silvio |
Advisor | Ermon, Stefano |
Advisor | Savarese, Silvio |
Subjects
Genre | Theses |
---|
Bibliographic information
Statement of responsibility | Jonathan Krause. |
---|---|
Note | Submitted to the Department of Computer Science. |
Thesis | Thesis (Ph.D.)--Stanford University, 2016. |
Location | electronic resource |
Access conditions
- Copyright
- © 2016 by Jonathan David Krause
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...