自動運転知覚のための自己教師あり学習:解析手法と学習手法の融合
Self-supervised learning for autonomous vehicles perception: A conciliation between analytical and learning methods
自動運転の知覚タスクにおいて、手動ラベリングを自動化する自己教師あり学習の概念を整理し、一般的な定式化と今後の課題を示したチュートリアル論文。
著者: Florent Chiaroni, Mohamed-Cherif Rahal, Nicolas Hueber, Frederic Dufaux
分類: cs.CV, cs.LG, cs.RO, eess.IV
原文アブストラクト
Nowadays, supervised deep learning techniques yield the best state-of-the-art prediction performances for a wide variety of computer vision tasks. However, such supervised techniques generally require a large amount of manually labeled training data. In the context of autonomous vehicles perception, this requirement is critical, as the distribution of sensor data can continuously change and include several unexpected variations. It turns out that a category of learning techniques, referred to as self-supervised learning (SSL), consists of replacing the manual labeling effort by an automatic labeling process. Thanks to their ability to learn on the application time and in varying environments, state-of-the-art SSL techniques provide a valid alternative to supervised learning for a variety of different tasks, including long-range traversable area segmentation, moving obstacle instance segmentation, long-term moving obstacle tracking, or depth map prediction. In this tutorial-style article, we present an overview and a general formalization of the concept of self-supervised learning (SSL) for autonomous vehicles perception. This formalization provides helpful guidelines for developing novel frameworks based on generic SSL principles. Moreover, it enables to point out significant challenges in the design of future SSL systems.
関連論文
- Vernata: LiDAR点群表現の自己教師あり学習自己教師あり学習
- マルチモダリティを教師とする自己教師あり学習:テスト環境への特化自己教師あり学習
- HilDA: 拡散モデルを用いた階層的蒸留による自己教師ありLiDAR事前学習の推進自己教師あり学習
- OctoSense: マルチモーダルロボット知覚のための自己教師あり学習自己教師あり学習
- 構造的不変性からの自己教師あり学習自己教師あり学習
- 海底画像に対する位置情報正則化自己教師あり特徴学習の検討自己教師あり学習
