語義分割
對圖像中每一個像素進行語義級分類，上圖人物是紅色，車輛是藍色，樹木是綠色，建築是灰色等。不區分不同的帶有相同語義的個體。因爲語義分割需要進行像素級分類，所以標註成本很高。
論文列表
語義分割（6篇）
Severity-Aware Semantic Segmentation With Reinforced Wasserstein Training
Squeeze-and-Attention Networks for Semantic Segmentation
Spatial Pyramid Based Graph Reasoning for Semantic Segmentation
Dual Super-Resolution Learning for Semantic Segmentation
Code：https://github.com/wanglixilinx/DSRL
Transferring and Regularizing Prediction for Semantic Segmentation
CascadePSP: Toward Class-Agnostic and Very High-Resolution Segmentation via Global and Local Refinement
Code: https://github.com/hkchengrex/CascadePSP
神經架構搜索（NAS）(1篇)
Graph-Guided Architecture Search for Real-Time Semantic Segmentation
動態網絡（Dynamic Networks）(Oral) (1篇)
Learning Dynamic Routing for Semantic Segmentation
Code：https://github.com/yanwei-li/DynamicRouting
視頻語義分割（1篇）
Temporally Distributed Networks for Fast Video Semantic Segmentation
Code： https://github.com/feinanshan/TDNet
