Fig. 1
YOLOX-NR model structure diagram
Download original image
(a) The Backbone of YOLOX-NR is based on CSPDarknet, with three added ECA attention block. C3, C4, and C5 are feature maps of different scales extracted by the Backbone network. (b)The Neck uses the structure of the PANet and is optimized by using ECA attention block. (c) Decoupled head is the detection of YOLOX-NR, with added α-IoU loss function. H3, H4, H5 are three output branches of different scales
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.