Recently, IEEE Signal Processing Society has just announced the awardee's list for "top 25 downloaded articles in 2020 for IEEE Transactions on Image Processing on IEEE Xplore". It is our great pleasure to know that our work entitled "Occlusion Aware Facial Expression Recognition Using CNN With Attention Mechanism" (Authors: Yong Li, Jiabei Zeng, Shiguang Shan, and Xilin Chen) has been identified as being one of them! Congratulations!
This paper proposes a convolution neutral network (CNN) with attention mechanism (ACNN) that can perceive the occlusion regions of the face and focus on the most discriminative un-occluded regions. The proposed ACNNs are evaluated on both real and synthetic occlusions, including a self-collected facial expression dataset with real-world occlusions, the two largest in-the-wild facial expression datasets and their modifications with synthesized facial occlusions. Experimental results show that ACNNs improve the recognition accuracy on both the non-occluded faces and occluded faces.
Download: