LanePerf: a Performance Estimation Framework for Lane Detection

TL;DR

This paper introduces LanePerf, a novel framework for estimating lane detection model performance without ground-truth labels, using image and lane features to improve robustness assessment across diverse driving conditions.

Contribution

The paper adapts existing classification performance estimation methods to lane detection and proposes LanePerf, a new framework that effectively handles zero-lane detection and significant domain shifts.

Findings

LanePerf outperforms baselines with lower MAE and higher correlation.

Extensive experiments on OpenLane show robustness across diverse conditions.

The framework supports safer, label-free performance estimation in ADAS.

Abstract

Lane detection is a critical component of Advanced Driver-Assistance Systems (ADAS) and Automated Driving System (ADS), providing essential spatial information for lateral control. However, domain shifts often undermine model reliability when deployed in new environments. Ensuring the robustness and safety of lane detection models typically requires collecting and annotating target domain data, which is resource-intensive. Estimating model performance without ground-truth labels offers a promising alternative for efficient robustness assessment, yet remains underexplored in lane detection. While previous work has addressed performance estimation in image classification, these methods are not directly applicable to lane detection tasks. This paper first adapts five well-performing performance estimation methods from image classification to lane detection, building a baseline. Addressing the limitations of prior approaches that solely rely on softmax scores or lane features, we further propose a new Lane Performance Estimation Framework (LanePerf), which integrates image and lane features using a pretrained image encoder and a DeepSets-based architecture, effectively handling zero-lane detection scenarios and large domain-shift cases. Extensive experiments on the OpenLane dataset, covering diverse domain shifts (scenes, weather, hours), demonstrate that our LanePerf outperforms all baselines, achieving a lower MAE of 0.117 and a higher Spearman's rank correlation coefficient of 0.727. These findings pave the way for robust, label-free performance estimation in ADAS, supporting more efficient testing and improved safety in challenging driving scenarios.