The Bridge - Issue 2, 2023 - 22

Graduate Research SP
Can Peng
Mu Kappa
The University of Queensland, Ph.D. Student in Computer Science
RESEARCH TOPIC
Computer Vision, Incremental Learning, Object Detection
Can pursued her Ph.D. at the University of Queensland (UQ). Prior to her Ph.D., she received her Bachelor in Electrical and Biomedical
Engineering at UQ. Can's Ph.D. thesis explores incremental learning on image classification and object detection tasks. Image classification
and object detection are significant tasks in the computer vision area and have been adopted in many real-life applications, such as
autopilot, augmented reality, and mobile shopping. In the dynamic real world, new objects of interest might continually appear over time.
Learning to detect and classify new objects incrementally in an accurate and timely
manner is crucial in many application scenarios. Unfortunately, current deep learning
networks are ill-equipped for handling incrementally appearing non-stationary
data distributions. The purpose of Can's Ph.D. study is to tackle dynamic real-world
scenarios and design effective incremental models that can continually learn new
tasks without forgetting the previously learned knowledge. After graduating from
UQ recently, she is interested in continuing to explore in the AI areas, especially in
the lifelong learning and embodied AI area. Can joined the Queensland Chapter of
the IEEE-HKN Society in 2021. During her Ph.D. life, besides the accumulation of
academic aspects, by participating in various interesting activities hosted by the HKN
chapter, she also strives to grow into an all-rounded person.
LEARN MORE
https://scholar.google.com.au/citations?user=HmzYOu4AAAAJ&hl=en
TLIGHT
An example of an incremental object detection task. Incremental
learning tasks require the model to continually learn new tasks
without forgetting the previously learned knowledge.
THE BRIDGE

https://scholar.google.com.au/citations?user=HmzYOu4AAAAJ&hl=en https://careers.ti.com/

The Bridge - Issue 2, 2023

Table of Contents for the Digital Edition of The Bridge - Issue 2, 2023