Design of A Two-Branch EfficientNet for Person Re-Identification

Due to its strong expressive ability, ResNet is widely used in Person Re-Identification. Although the use of ResNet50 for person ReID has achieved excellent performance, the popular ResNet-50-based solution consumes considerable resources with low efficiency in implementations. Recently, EfficientNet was proposed as an emerging alternative for design of deep neural models with much better efficiency. In this paper, we propose to design a two-branch EfficientNet for person ReID. Compared with various ResNet-50-based solutions, the proposed EfficientNet has a significantly small size on network, but often achieves better performance. Experimental results show that the proposed network performs very well on the popular Person Re-Identification data set.