Skip to main content
EURASIP Journal on Wireless Communications and Networking
Fig. 3 | EURASIP Journal on Wireless Communications and Networking

Fig. 3

From: Tree-CNN: from generalization to specialization

Fig. 3

The flow chart of obtaining the coarse categories. Firstly, Trunk-NET generates the probability of every training image. Secondly, Eq. 1 generates similarity map with the probability of every training image. Thirdly, according to the similarity map, Algorithm 1 engenders coarse categories with threshold T

Back to article page