Synthetic-to-Real Domain Adaptation for Lane Detection

TL;DR

This paper presents a novel autoencoder-based domain adaptation method for lane detection that effectively leverages synthetic data and minimal labeled real data, reducing the need for extensive manual annotation.

Contribution

The work introduces a new autoencoder approach for unsupervised and semi-supervised domain adaptation in lane detection, outperforming existing methods.

Findings

Autoencoder approach nearly matches fully supervised accuracy with only 10% labeled data.

Method outperforms other domain adaptation techniques in semi-supervised setting.

Effective in reducing costly target domain labeling efforts.

Abstract

Accurate lane detection, a crucial enabler for autonomous driving, currently relies on obtaining a large and diverse labeled training dataset. In this work, we explore learning from abundant, randomly generated synthetic data, together with unlabeled or partially labeled target domain data, instead. Randomly generated synthetic data has the advantage of controlled variability in the lane geometry and lighting, but it is limited in terms of photo-realism. This poses the challenge of adapting models learned on the unrealistic synthetic domain to real images. To this end we develop a novel autoencoder-based approach that uses synthetic labels unaligned with particular images for adapting to target domain data. In addition, we explore existing domain adaptation approaches, such as image translation and self-supervision, and adjust them to the lane detection task. We test all approaches in the unsupervised domain adaptation setting in which no target domain labels are available and in the semi-supervised setting in which a small portion of the target images are labeled. In extensive experiments using three different datasets, we demonstrate the possibility to save costly target domain labeling efforts. For example, using our proposed autoencoder approach on the llamas and tuSimple lane datasets, we can almost recover the fully supervised accuracy with only 10% of the labeled data. In addition, our autoencoder approach outperforms all other methods in the semi-supervised domain adaptation scenario.