• English
    • Tiếng Việt
  • English 
    • English
    • Tiếng Việt
  • Login
View Item 
  •   VinSpace Home
  • The College of Engineering and Computer Science
  • Pham Huy Hieu, PhD.
  • View Item
  •   VinSpace Home
  • The College of Engineering and Computer Science
  • Pham Huy Hieu, PhD.
  • View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.

Multi-stream fusion for class incremental learning in pill image classification

Thumbnail
View/Open
Multi-stream Fusion for Class Incremental Learning in Pill Image Classification.pdf (1.183Mb)
Date
2022
Author
Pham, Huy Hieu
Nguyen, Trong Tung
Nguyen, Phi Le
Nguyen, Thanh Hung
Do, Minh
Metadata
Show full item record
Abstract
Classifying pill categories from real-world images is crucial for various smart healthcare applications. Although existing approaches in image classification might achieve a good performance on fixed pill categories, they fail to handle novel instances of pill categories that are frequently presented to the learning algorithm. To this end, a trivial solution is to train the model with novel classes. However, this may result in a phenomenon known as catastrophic forgetting, in which the system forgets what it learned in previous classes. In this paper, we address this challenge by introducing the class incremental learning (CIL) ability to traditional pill image classification systems. Specifically, we propose a novel incremental multi-stream intermediate fusion framework enabling incorporation of an additional guidance information stream that best matches the domain of the problem into various state-of-the-art CIL methods. From this framework, we consider color-specific information of pill images as a guidance stream and devise an approach, namely “Color Guidance with Multi-stream intermediate fusion” (CG-IMIF) for solving CIL pill image classification task. We conduct comprehensive experiments on real-world incremental pill image classification dataset, namely VAIPE-PCIL, and find that the CG-IMIF consistently outperforms several state-of-the-art methods by a large margin in different task settings. Our code, data, and trained model are available at https://github.com/vinuni-vishc/CG-IMIF.
URI
https://vinspace.edu.vn/handle/VIN/318
Collections
  • Pham Huy Hieu, PhD. [36]

Contact Us | Send Feedback
 

 

Browse

All of VinSpaceCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

My Account

LoginRegister

Contact Us | Send Feedback