Spatial As Deep: Spatial CNN for Traffic Scene Understanding

17 Dec 2017Xingang PanJianping ShiPing LuoXiaogang WangXiaoou Tang

Convolutional neural networks (CNNs) are usually built by stacking convolutional operations layer-by-layer. Although CNN has shown strong capability to extract semantics from raw pixels, its capacity to capture spatial relationships of pixels across rows and columns of an image is not fully explored... (read more)

PDF Abstract

Evaluation Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK COMPARE
Lane Detection CULane SCNN F1 score 71.6 # 1
Lane Detection TuSimple Spatial CNN Accuracy 96.53% # 2