Learning visual concepts with fewer human annotations

Thesis

This thesis explores the use of modern deep neural networks to learn visual concepts with fewer human annotations on data. While data is abundant and increasingly easier to collect, most deep learning methods need extensive human labelling to be trained, which is often costly and may require expert-level knowledge. In this thesis we explore alternatives to human labelling by considering synthetic data, as well as partially and completely unlabelled data. We will study these alternatives wi...

Expand abstract

Authors: Ehrhardt, S

• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford
• Language: English
• Keywords: Object Recognition; Machine Learning; Weakly-supervised Learning; Unsupervised Learning; Computer Vision; Artificial Intelligence; Neural Physics
• Subjects: Machine learning; Computer vision; Weakly-supervised learning; Artifical Intelligence