A Deep Learning-Based Semantic Segmentation Architecture for Autonomous Driving Applications

Table 4

Experimental results of segmentation models on Mapillary Vistas.

Model	Number of parameters	Size	mIoU
ENet	688 K	3 Mb	56
M-Net+ASPP	375 K	1.5 Mb	61
M-Net+PSP	348 K	1.38 Mb	59
SINet	43 K	0.3 Mb	50