制御のための視覚的3Dキーポイントの教師なし学習
Unsupervised Learning of Visual 3D Keypoints for Control
画像から微分可能なエンコーダで3Dキーポイントを教師なし学習し、多視点一貫性と制御タスク目的を同時最適化することで、ロボット関節や物体運動を捉え強化学習ベンチマークで従来手法を上回った。
著者: Boyuan Chen, Pieter Abbeel, Deepak Pathak
分類: cs.LG, cs.CV, cs.RO
原文アブストラクト
Learning sensorimotor control policies from high-dimensional images crucially relies on the quality of the underlying visual representations. Prior works show that structured latent space such as visual keypoints often outperforms unstructured representations for robotic control. However, most of these representations, whether structured or unstructured are learned in a 2D space even though the control tasks are usually performed in a 3D environment. In this work, we propose a framework to learn such a 3D geometric structure directly from images in an end-to-end unsupervised manner. The input images are embedded into latent 3D keypoints via a differentiable encoder which is trained to optimize both a multi-view consistency loss and downstream task objective. These discovered 3D keypoints tend to meaningfully capture robot joints as well as object movements in a consistent manner across both time and 3D space. The proposed approach outperforms prior state-of-art methods across a variety of reinforcement learning benchmarks. Code and videos at https://buoyancy99.github.io/unsup-3d-keypoints/
