Homepage

Looking for Job Opportunity: I am looking for PostDoc/Assi. Prof./Assi. Researcher opportunities. Kindly let me know if I am suited to the job and you are interested in working with me. Many thanks :)

Mentorship & Collaboration: From times to times, I mentor students and am passionate to collaborate with people on amazing research projects/ideas. If you would like to collaborate with me, don’t hesitate to drop me an email.

Changsheng Lu [CV-English, CV-中文] is anticipated to obtain his PhD degree from the College of Engineering and Computer Science at The Australian National University in July 2024. His PhD advisor is Dr. Piotr Koniusz. Previously, he received the M.S. degree in control science and engineering from Shanghai Jiao Tong University, Shanghai, China, in 2020, and received the B.S. degree in Automation from Southeast University, Nanjing, China, in 2017. He has wide research interests across computer vision, machine learning, artificial intelligence, and robotics. Particularly, he is interested in the theories and algorithms that may improve a robot or an intelligent device’s peception, cognition and reasoning ability, which could empower the machine to see, think and conduct more like a human.

Currently, he is focusing on the research topics in multimodal generic vision, keypoint detection, zero- and few-shot learning, foundation model, and transfer learning.

[go top]

News

Two papers are accepted to ECCV 2024! One paper is about zero- and few-shot keypoint detection (Z-FSKD) (one of my best works); another one is about motion transfer with realistic apparel animation co-authored with my colleague Rong Wang. I will attend ECCV 2024 in Milano, Italy.
Our paper ``Detect Any Keypoints: An Efficient Light-Weight Few-Shot Keypoint Detector’’ has been accepted to AAAI’24 (the 38th Annual AAAI Conference on Artificial Intelligence)!
One co-first authored paper about Few-shot shape recognition (FSSD) has been accepted to WACV’24. To our best knowledge, this is the first work in the field.
Two co-authored papers (Re. Diatom object detection & Shadow removal) have been accepted to IJCNN’23.
One research paper about Few-shot Keypoint Detection (FSKD) gets accepted by CVPR’22.
Two co-authored papers get accepted by ACCV’22 (Re. Deep General Ellipse detection) and ICONIP’22 (Re. Fast Stain Color Normalization). One co-authored paper about industrial scene text detection gets accepted by IEEE TCSVT.
Two papers about domain adaptation/transfer learning published in ICONIP’18 and Neurocomputing. [here]
Our High-quality Ellipse Detector gets accepted by IEEE T-IP and exceeds 450+ stars in github (rank top1 so far). We also have an IEEE ICIP version for circle detection.