Large-scale labeled datasets are generally required to train deep neural networks in order to obtain better performance in visual feature learning for computer vision applications. To reduce the extensive cost of collecting and annotating large-scale labeled datasets, various machine learning methods are proposed to learn general visual features including semi-supervised methods which learn visual features from a small size of labeled data and a large amount of unlabeled data, weakly supervised methods which learn visual features from coarse-grained labeled data, and self-supervised methods which learn visual features from large-scale unlabeled data. In this thesis, we investigate a number of approaches to learn robust deep visual features from data with different level of supervisions including a weakly supervised method, a semi-supervised learning method, and several self-supervised learning methods. To demonstrate the generalization ability of the proposed methods to learn from limited supervisions, we validate the proposed methods on different tasks and demonstrate that the proposed methods indeed can learn robust visual features from limited labeled data.
","abstract_html":"<p>Large-scale labeled datasets are generally required to train deep neural networks in order to obtain better performance in visual feature learning for computer vision applications. To reduce the extensive cost of collecting and annotating large-scale labeled datasets, various machine learning methods are proposed to learn general visual features including semi-supervised methods which learn visual features from a small size of labeled data and a large amount of unlabeled data, weakly supervised methods which learn visual features from coarse-grained labeled data, and self-supervised methods which learn visual features from large-scale unlabeled data. In this thesis, we investigate a number of approaches to learn robust deep visual features from data with different level of supervisions including a weakly supervised method, a semi-supervised learning method, and several self-supervised learning methods. To demonstrate the generalization ability of the proposed methods to learn from limited supervisions, we validate the proposed methods on different tasks and demonstrate that the proposed methods indeed can learn robust visual features from limited labeled data.</p>","abstract_has_math":false,"creators":["Jing, Longlong"],"institution":"The Graduate School and University Center of The City University of New York","degree_name":"Doctor of Philosophy","degree_level":"Doctoral","degree_discipline":"Computer Science","degree_department":null,"school":null,"contributors":[],"advisors":["Yingli Tian"],"committee_chairs":[],"committee_members":["Zhigang Zhu","Ioannis Stamos","Charles Qi"],"year":2021,"date_issued":"2021-09-01T07:00:00Z","date_published":"2021-09-01T07:00:00Z","updated_at":"2026-07-24T01:59:05Z","subjects":["computer vision","deep learning"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://academicworks.cuny.edu/gc_etds/4635","outbound_label":"Repository record","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Yingli Tian"]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Zhigang Zhu","Ioannis Stamos","Charles Qi"]},{"key":"dc:creator","label":"Author","values":["Jing, Longlong"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2022-09-30T07:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Doctoral"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Doctor of Philosophy"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["The Graduate School and University Center of The City University of New York"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["computer vision","deep learning"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://academicworks.cuny.edu/gc_etds/4635"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Large-scale labeled datasets are generally required to train deep neural networks in order to obtain better performance in visual feature learning for computer vision applications. To reduce the extensive cost of collecting and annotating large-scale labeled datasets, various machine learning methods are proposed to learn general visual features including semi-supervised methods which learn visual features from a small size of labeled data and a large amount of unlabeled data, weakly supervised methods which learn visual features from coarse-grained labeled data, and self-supervised methods which learn visual features from large-scale unlabeled data. In this thesis, we investigate a number of approaches to learn robust deep visual features from data with different level of supervisions including a weakly supervised method, a semi-supervised learning method, and several self-supervised learning methods. To demonstrate the generalization ability of the proposed methods to learn from limited supervisions, we validate the proposed methods on different tasks and demonstrate that the proposed methods indeed can learn robust visual features from limited labeled data.
"]},{"key":"dc:title","label":"Title","values":["Learning Deep Visual Features from Limited Labeled Data"]}]}],"canonical_facts":{"dc:contributor.advisor":["Yingli Tian"],"dc:contributor.committeemember":["Zhigang Zhu","Ioannis Stamos","Charles Qi"],"dc:creator":["Jing, Longlong"],"dc:date.available":["2022-09-30T07:00:00Z"],"dc:description.abstract":["Large-scale labeled datasets are generally required to train deep neural networks in order to obtain better performance in visual feature learning for computer vision applications. To reduce the extensive cost of collecting and annotating large-scale labeled datasets, various machine learning methods are proposed to learn general visual features including semi-supervised methods which learn visual features from a small size of labeled data and a large amount of unlabeled data, weakly supervised methods which learn visual features from coarse-grained labeled data, and self-supervised methods which learn visual features from large-scale unlabeled data. In this thesis, we investigate a number of approaches to learn robust deep visual features from data with different level of supervisions including a weakly supervised method, a semi-supervised learning method, and several self-supervised learning methods. To demonstrate the generalization ability of the proposed methods to learn from limited supervisions, we validate the proposed methods on different tasks and demonstrate that the proposed methods indeed can learn robust visual features from limited labeled data.
"],"dc:identifier":["https://academicworks.cuny.edu/gc_etds/4635"],"dc:subject":["computer vision","deep learning"],"dc:title":["Learning Deep Visual Features from Limited Labeled Data"],"thesis:degree_discipline":["Computer Science"],"thesis:degree_level":["Doctoral"],"thesis:degree_name":["Doctor of Philosophy"],"thesis:institution_name":["The Graduate School and University Center of The City University of New York"]},"updated_at":"2026-07-24T01:59:05Z"}