距離センサと衛星画像間の自己教師あり位置推定
Self-Supervised Localisation between Range Sensors and Overhead Imagery
地上の距離センサデータと衛星画像という異なるモダリティ間の位置合わせを、正解位置情報なしの自己教師あり学習で実現する手法を提案し、特にミリ波レーダでの有効性を示した。
著者: Tim Y. Tang, Daniele De Martini, Shangzhe Wu, Paul Newman
分類: cs.RO, cs.CV
原文アブストラクト
Publicly available satellite imagery can be an ubiquitous, cheap, and powerful tool for vehicle localisation when a prior sensor map is unavailable. However, satellite images are not directly comparable to data from ground range sensors because of their starkly different modalities. We present a learned metric localisation method that not only handles the modality difference, but is cheap to train, learning in a self-supervised fashion without metrically accurate ground truth. By evaluating across multiple real-world datasets, we demonstrate the robustness and versatility of our method for various sensor configurations. We pay particular attention to the use of millimetre wave radar, which, owing to its complex interaction with the scene and its immunity to weather and lighting, makes for a compelling and valuable use case.
