On the Robustness and Generalization of Object Detection and Game Applications by Dr. Siqi Yang(Online)
Speaker: Dr. Siqi Yang
Topic: On the Robustness and Generalization of Object Detection and Game Applications
Time & Date: 10:00 -11:00, Friday,December 30, 2022 (Beijing time)
Zoom Meeting:
https://cuhk
Zoom Meeting ID:993 1113 2075
Password：123456
Abstract:
Object detection is to recognize and locate objects in images, which is one of the fundamental problems in computer vision. It has been developed rapidly in the past decades and has a wide range of applications, e.g., surveillance systems, autonomous vehicles, medical imaging diagnosis, and game applications. Since the technique of object detection has been widely adopted, the lack of robustness of object detectors can be problematic, which may lead to unpredictable safety and financial losses. The main goal of this thesis is to study the robustness of object detectors and develop robust detectors. Robust object detectors can be viewed as the detectors that have the ability to maintain high accuracy when encountering the challenging conditions that may degrade the performance. To develop robust object detectors, we study the robustness of object detection by understanding when and where the detectors may fail. In this talk, we focus on the failure cases in three aspects: false positives, attacks, and domain adaptability. The proposed domain adaptation techniques have been applied to medical and game applications.
Biography:
Siqi Yang is currently a research scientist at the Innovation and Entertainment Group of Tencent, working on conditional motion generation for virtual humans. Prior to joining Tencent, she was a research scientist at the Game AI Lab of Bytedance, where she worked on game character auto-creation, image generation and other computer vision application in game. Siqi earned her Ph.D. degree in computer science from the University of Queensland, Australia in 2020 and BEng degree from South China University of Technology. Her research interests are machine learning and computer vision, particularly transfer learning, domain adaptation, generative networks, adversarial robustness, object detection and medical image processing. She has published 9 papers on top-tier computer vision conferences and 4 granted patents.
