VDD: Varied Drone Dataset for Semantic Segmentation

TL;DR

The paper introduces VDD, a large-scale, diverse drone image dataset with high-resolution images across multiple scenes and classes, aiming to improve semantic segmentation accuracy in aerial drone imagery.

Contribution

It provides a new, comprehensive drone dataset with diverse scenes and annotations, addressing limitations of existing datasets and facilitating advancements in drone image segmentation.

Findings

Seven state-of-the-art models trained on VDD serve as baselines.

VDD covers urban, industrial, rural, and natural scenes.

Public availability of the dataset encourages further research.

Abstract

Semantic segmentation of drone images is critical for various aerial vision tasks as it provides essential semantic details to understand scenes on the ground. Ensuring high accuracy of semantic segmentation models for drones requires access to diverse, large-scale, and high-resolution datasets, which are often scarce in the field of aerial image processing. While existing datasets typically focus on urban scenes and are relatively small, our Varied Drone Dataset (VDD) addresses these limitations by offering a large-scale, densely labeled collection of 400 high-resolution images spanning 7 classes. This dataset features various scenes in urban, industrial, rural, and natural areas, captured from different camera angles and under diverse lighting conditions. We also make new annotations to UDD and UAVid, integrating them under VDD annotation standards, to create the Integrated Drone Dataset (IDD). We train seven state-of-the-art models on drone datasets as baselines. It's expected that our dataset will generate considerable interest in drone image segmentation and serve as a foundation for other drone vision tasks. Datasets are publicly available at \href{our website}{https://github.com/RussRobin/VDD}.