Learning to Discover Multi-Class Attentional Regions for Multi-Label Image Recognition

• Published in: IEEE transactions on image processing Vol. 30; pp. 5920 - 5932
• Main Authors: Gao, Bin-Bin, Zhou, Hong-Yu
• Format: Journal Article
• Language: English
• Published: New York IEEE 2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: attentional region; Computer architecture; global to local; Image classification; Image recognition; Modules; multi-class; Multi-label; Object recognition; Proposals; Reinforcement learning; Semantics; Streaming media; Task analysis; two-stream; Visualization

Multi-label image recognition is a practical and challenging task compared to single-label image classification. However, previous works may be suboptimal because of a great number of object proposals or complex attentional region generation modules. In this paper, we propose a simple but efficient...