変換整合性損失を用いた教師なしメトリック再位置推定
Unsupervised Metric Relocalization Using Transform Consistency Loss
地図内でクエリ画像を位置推定する際、参照画像によらず同じ絶対姿勢が得られるべきという性質を利用し、変換整合性損失で深層ネットワークを自己教師あり学習させ、動的環境でのロバストなメトリック再位置推定を実現した。
著者: Mike Kasper, Fernando Nobre, Christoffer Heckman, Nima Keivan
分類: cs.CV, cs.RO
原文アブストラクト
Training networks to perform metric relocalization traditionally requires accurate image correspondences. In practice, these are obtained by restricting domain coverage, employing additional sensors, or capturing large multi-view datasets. We instead propose a self-supervised solution, which exploits a key insight: localizing a query image within a map should yield the same absolute pose, regardless of the reference image used for registration. Guided by this intuition, we derive a novel transform consistency loss. Using this loss function, we train a deep neural network to infer dense feature and saliency maps to perform robust metric relocalization in dynamic environments. We evaluate our framework on synthetic and real-world data, showing our approach outperforms other supervised methods when a limited amount of ground-truth information is available.
