Shaogang (Sean) Gong is Professor of Visual Computation at Queen Mary University of London and a Turing Fellow of the Alan Turing Institute of Data Science and Artificial Intelligence. He established the Queen Mary Computer Vision Laboratory in 1993 and has enjoyed immensely working with PhD students and postdoctoral researchers. His research is in Computer Vision and Machine Learning (Google Scholar and DBLP), with a focus on Object Recognition, Action Recognition, and Video Analysis. His brief bio and publications with pdf download.
SEARCH AND LEARN: FROM USER GUIDED SEARCH TO FEDERATED ZERO-SHOT LEARNING
Deep learning has been successful for many computer vision tasks due to the availability of shared and centralised large sized training data. However, increasing awareness of privacy concerns poses new challenges to deep learning, especially for human subject related recognition such as person reidentification (Re-ID). Moreover, existing person search methods predominantly assume the availability of at least one-shot imagery sample of the queried person. This assumption is limited in circumstances where only a brief textual (or verbal) description of the target person is available.
In this talk, I will describe challenges and recent progress on deep learning for text attribute based person search without any query image, and decentralised learning from non-shared private training data distributed at multiple user-cites of independent multidomain labels for person re-identification. Both problems require solving generalised Zero-Shot Learning.