Learning Lightweight Lane Detection CNNs by Self Attention Distillation

Abstract

Training deep models for lane detection is challenging due to the very subtle and sparse supervisory signals inherent in lane annotations. Without learning from much richer context, these models often fail in challenging scenarios, e.g., severe occlusion, ambiguous lanes, and poor lighting conditions. In this paper, we present a novel knowledge distillation approach, i.e., Self Attention Distillation (SAD), which allows a model to learn from itself and gains substantial improvement without any additional supervision or labels. Specifically, we observe that attention maps extracted from a model trained to a reasonable level would encode rich contextual information. The valuable contextual information can be used as a form of 'free' supervision for further representation learning through performing topdown and layer-wise attention distillation within the network itself. SAD can be easily incorporated in any feedforward convolutional neural networks (CNN) and does not increase the inference time. We validate SAD on three popular lane detection benchmarks (TuSimple, CULane and BDD100K) using lightweight models such as ENet, ResNet-18 and ResNet-34. The lightest model, ENet-SAD, performs comparatively or even surpasses existing algorithms. Notably, ENet-SAD has 20 x fewer parameters and runs 10 x faster compared to the state-of-the-art SCNN, while still achieving compelling performance in all benchmarks. Our code is available at https://github.com/cardwing/Codes-for-Lane-Detection.

arXiv comment9 pages, 8 figures; This paper is accepted by ICCV 2019; Our code is available at https://github.com/cardwing/Codes-for-Lane-Detection

References in corpus (4)

• Distilling the Knowledge in a Neural Network
• Paying More Attention to Attention: Improving the Performance of Convolutional Neural Networks via Attention Transfer
• ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation
• Spatial As Deep: Spatial CNN for Traffic Scene Understanding

Cited by in corpus (9)

• Knowledge Distillation: A Survey
• Collaborative Distillation in the Parameter and Spectrum Domains for Video Action Recognition
• Spatial Likelihood Voting with Self-Knowledge Distillation for Weakly Supervised Object Detection
• Flexible 3D Lane Detection by Hierarchical Shape MatchingFlexible 3D Lane Detection by Hierarchical Shape Matching
• On Self-Distilling Graph Neural Network
• Model-Agnostic Defense for Lane Detection against Adversarial Attack
• Domain-Agnostic Clustering with Self-Distillation
• WeClick: Weakly-Supervised Video Semantic Segmentation with Click Annotations
• DAGMapper: Learning to Map by Discovering Lane Topology