Class-Level Feature Disentanglement for Multi-Label Image Classification

Yingduo Tong, Zhenyu Lu, Yize Dong and Yonggang Lu ()
Additional contact information
Yingduo Tong: School of Information Science and Engineering, Lanzhou University, Lanzhou 730000, China
Zhenyu Lu: School of Educational Technology, NorthWest Normal University, Lanzhou 730070, China
Yize Dong: School of Information Science and Engineering, Lanzhou University, Lanzhou 730000, China
Yonggang Lu: School of Information Science and Engineering, Lanzhou University, Lanzhou 730000, China

Future Internet, 2025, vol. 17, issue 11, 1-17

Abstract: Generally, the interpretability of deep neural networks is categorized into a priori and a posteriori interpretability. A priori interpretability involves improving model transparency through deliberate design prior to training. Feature disentanglement is a method for achieving a priori interpretability. Existing disentanglement methods mostly focus on semantic features, such as intrinsic and shared features. These methods distinguish between the background and the main subject, but overlook class-level features in images. To address this, we take a further step by advancing feature disentanglement to the class level. For multi-label image classification tasks, we propose a class-level feature disentanglement method. Specifically, we introduce a multi-head classifier within the feature extraction layer of the backbone network to disentangle features. Each head in this classifier corresponds to a specific class and generates independent predictions, thereby guiding the model to better leverage the intrinsic features of each class and improving multi-label classification precision. Experiments demonstrate that our method significantly enhances performance metrics across various benchmarks while simultaneously achieving a priori interpretability.

Keywords: priori interpretability; feature disentanglement; multi-head classifier; image classification (search for similar items in EconPapers)
JEL-codes: O3 (search for similar items in EconPapers)
Date: 2025
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/1999-5903/17/11/486/pdf (application/pdf)
https://www.mdpi.com/1999-5903/17/11/486/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jftint:v:17:y:2025:i:11:p:486-:d:1778133

Access Statistics for this article

Future Internet is currently edited by Ms. Grace You

More articles in Future Internet from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().