Inter-region affinity distillation for road marking segmentation

Abstract

We study the problem of distilling knowledge from a large deep teacher network to a much smaller student net- work for the task of road marking segmentation. In this work, we explore a novel knowledge distillation (KD) ap- proach that can transfer ‘knowledge’ on scene structure more effectively from a teacher to a student model. Our method is known as Inter-Region Affinity KD (IntRA-KD). It decomposes a given road scene image into different re- gions and represents each region as a node in a graph. An inter-region affinity graph is then formed by establishing pairwise relationships between nodes based on their sim- ilarity in feature distribution. To learn structural knowl- edge from the teacher network, the student is required to match the graph generated by the teacher. The proposed method shows promising results on three large-scale road marking segmentation benchmarks, i.e., ApolloScape, CU- Lane and LLAMAS, by taking various lightweight mod- els as students and ResNet-101 as the teacher. IntRA- KD consistently brings higher performance gains on all lightweight models, compared to previous distillation meth- ods. Our code is available at https://github.com/ cardwing/Codes-for-IntRA-KD.