PaRK-Detect: Towards Efficient Multi-Task Satellite Imagery Road Extraction via Patch-Wise Keypoints Detection

TL;DR

PaRK-Detect introduces a multi-task, patch-wise keypoints detection method for satellite road extraction, achieving high accuracy and faster inference by combining keypoint detection with semantic segmentation.

Contribution

The paper presents a novel multi-task framework that predicts patch-wise road keypoints and their relationships, improving efficiency and connectivity over existing methods.

Findings

Achieves competitive or superior accuracy on benchmark datasets.

Demonstrates significantly faster inference speed.

Effectively constructs road graphs in a single pass.

Abstract

Automatically extracting roads from satellite imagery is a fundamental yet challenging computer vision task in the field of remote sensing. Pixel-wise semantic segmentation-based approaches and graph-based approaches are two prevailing schemes. However, prior works show the imperfections that semantic segmentation-based approaches yield road graphs with low connectivity, while graph-based methods with iterative exploring paradigms and smaller receptive fields focus more on local information and are also time-consuming. In this paper, we propose a new scheme for multi-task satellite imagery road extraction, Patch-wise Road Keypoints Detection (PaRK-Detect). Building on top of D-LinkNet architecture and adopting the structure of keypoint detection, our framework predicts the position of patch-wise road keypoints and the adjacent relationships between them to construct road graphs in a single pass. Meanwhile, the multi-task framework also performs pixel-wise semantic segmentation and generates road segmentation masks. We evaluate our approach against the existing state-of-the-art methods on DeepGlobe, Massachusetts Roads, and RoadTracer datasets and achieve competitive or better results. We also demonstrate a considerable outperformance in terms of inference speed.