MANet: End-to-End Learning for Point Cloud Based on Robust Pointpillar and Multiattention

Figure 7

The loss of train. (a) The ablation experiments of different attention. (b) The component of the total loss.

(a)

(b)