2024 3rd International Conference on Image Processing, Object Detection and Tracking(IPODT 2024)
History of IPODT 2022
Home / History of IPODT 2022



IPODT 2022

The Successful Conclusion of IPODT 2022

微信图片_202207161643214.png微信图片_202207161643211.png

2022 International Conference on Image Processing, Object Detection and Tracking(IPODT 2022) was organized by GSRA, Co-Organized byDalian University of Technology it was successfully held in the form of online conference on July 16, 2022.


Keynote Speech

2022-07-16_105224 (2).png

Prof. Shuai Wang,Shandong University, China

Title: Deep Learning for Medical Image Segmentation

Abstract: My title will be "Deep Learning for Medical Image Segmentation”, in which I will mainly talk about my research works based on deep learning in the field of medical image segmentation in recent years, including the conventional supervised-based methods and weakly supervised-based methods and their applications on the prostate segmentation and lymph node detection tasks. Specifically, for the conventional supervised-based methods, I will introduce some complex multi-task networks and for the weakly supervised-based methods, I will introduce some user friendly annotation-based methods.


2022-07-16_112354 (2).png

Prof. Lu Leng,Nanchang Hangkong University, China

Title: Cancelable Biometric

Abstract: Biometric recognition is convenient and reliable, so it has been widely used for identification and verification. However, the biometric systems typically suffer from many serious security and privacy problems. Biometric features are immutable, so they cannot be updated. In other words, a user cannot revoke and reissue his/her biometric template even if the template is compromised. In addition, a user’s biometric templates with the same features are likely to be stored and shared in various databases. If the user’s biometric template in one database is attacked successfully, his/her biometric templates in the other databases are not secure anymore. Last but not least, the biometric templates without protection likely leak users’ private information, such as gene defects, diseases. Cancelable biometric is a promising scheme for biometric protection, which should meet four criteria, namely diversity, revocability / reusability, non-invertibility, accuracy performance. Unfortunately, it is highly challenging to meet all the criteria simultaneously. This keynote speech will introduce the advanced technologies of cancelable biometric.


Oral Presentation

 2022-07-16_140628 (2).png


2022-07-16_141947 (2).png


2022-07-16_143153 (2).png