Awesome

deeplebv3plus-hedplus

Semantic segmentation model with the fusion of deeplabv3plus, hed and attention loss.

Rui Guo, Ronghua Liu, Na Li, Wei Liu

DV3+HED+: A DCNNs-based Framework to Monitor Temporary Works and ESAs in Railway Construction Project Using VHR Satellite Images

https://arxiv.org/abs/1908.11080

References:

[1]Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation

[2]Holistically-Nested Edge Detection

[3]DOOBNet: Deep Object Occlusion Boundary Detection from an Image