OMR: Occlusion-Aware Memory-Based Refinement for Video Lane Detection

TL;DR

This paper introduces OMR, an occlusion-aware memory-based refinement algorithm for video lane detection that improves accuracy by effectively handling occlusions and leveraging temporal information.

Contribution

The paper presents a novel occlusion-aware memory module and a data augmentation scheme, advancing video lane detection accuracy over existing methods.

Findings

Outperforms existing video lane detection techniques on benchmark datasets.

Effectively handles occlusions through a new mask detection and memory refinement process.

Demonstrates improved robustness and accuracy in complex driving scenarios.

Abstract

A novel algorithm for video lane detection is proposed in this paper. First, we extract a feature map for a current frame and detect a latent mask for obstacles occluding lanes. Then, we enhance the feature map by developing an occlusion-aware memory-based refinement (OMR) module. It takes the obstacle mask and feature map from the current frame, previous output, and memory information as input, and processes them recursively in a video. Moreover, we apply a novel data augmentation scheme for training the OMR module effectively. Experimental results show that the proposed algorithm outperforms existing techniques on video lane datasets. Our codes are available at https://github.com/dongkwonjin/OMR.