ADPNet: Attention based dual path network for lane detection

•We propose a dual path network with Detail Path and Semantic Path to capture complementary features to segment the input image into binary image indicating which pixels belong to lane or background.•We design a Dual Attention Module (DAM) in the Semantic Path. The DAM is composed of Position Attention Module and Channel Attention Module which can capture non-local relationships and refine the features for better representations in spatial and channel dimensions respectively.•The proposed architecture achieves comparable performance in terms of accuracy and real-time performance on the TuSimple and CULane lane detection benchmarks.