UAVによるセマンティックマッピングのための能動学習
Active Learning for UAV-based Semantic Mapping
深層学習の不確実性推定に基づく新規性評価を経路計画に組み込み、地形マッピング用の訓練データ収集を効率化するシステムを提案し、実データで従来の網羅的収集より大幅に少ないデータで済むことを示した。
著者: Hermann Blum, Silvan Rohrbach, Marija Popovic, Luca Bartolomei, Roland Siegwart
分類: cs.RO
原文アブストラクト
Unmanned aerial vehicles combined with computer vision systems, such as convolutional neural networks, offer a flexible and affordable solution for terrain monitoring, mapping, and detection tasks. However, a key challenge remains the collection and annotation of training data for the given sensors, application, and mission. We introduce an informative path planning system that incorporates novelty estimation into its objective function, based on research for uncertainty estimation in deep learning. The system is designed for data collection to reduce both the number of flights and of annotated images. We evaluate the approach on real world terrain mapping data and show significantly smaller collected training dataset compared to standard lawnmower data collection techniques.
