動きから物体を検出する弱教師あり学習手法
Towards Object Detection from Motion
物体が動く動画と物体のない負例動画の2種類だけで、位置アノテーションなしに物体検出器を学習する手法を提案し、ロボティクス設定で評価した。
著者: Rico Jonschkowski, Austin Stone
分類: cs.CV, cs.LG, cs.RO, stat.ML
原文アブストラクト
We present a novel approach to weakly supervised object detection. Instead of annotated images, our method only requires two short videos to learn to detect a new object: 1) a video of a moving object and 2) one or more "negative" videos of the scene without the object. The key idea of our algorithm is to train the object detector to produce physically plausible object motion when applied to the first video and to not detect anything in the second video. With this approach, our method learns to locate objects without any object location annotations. Once the model is trained, it performs object detection on single images. We evaluate our method in three robotics settings that afford learning objects from motion: observing moving objects, watching demonstrations of object manipulation, and physically interacting with objects (see a video summary at https://youtu.be/BH0Hv3zZG_4).
