器用なポイントポリシー：人間のデモから学習するポイントベースの器用な手のポリシー
Dexterous Point Policy: Learning Point-based Dexterous Hand Policies from Human Demonstrations
人間のビデオから直接、器用な操作ポリシーを学習するフレームワークを提案。3Dキーポイント表現で人間とロボットの身体性のギャップを埋め、ロボットデモなしで実ロボットタスクを高い成功率で達成。
詳しい要約
1. どんなもの？
2. 先行研究と比べてどこがすごい？
3. 技術・手法の肝は？
4. どうやって有効だと検証した？
5. 議論はある？
6. 次に読むべき論文は？
※ AIが要旨から生成した要約です。正確性は原文をご確認ください。
著者: Beomjun Kim, Seong Hyeon Park, Seunghoon Sim, Seungjun Moon, Sanghyeok Lee, Jinwoo Shin
分類: cs.RO, cs.CV, cs.LG
原文アブストラクト
Robotic foundation models pre-trained on human demonstration videos have shown promise, but a significant embodiment gap remains when the resulting policies are deployed on real robots. A common remedy is to fine-tune these models on robot-specific demonstrations. However, robot data collection can be prohibitively expensive and time-consuming, which is particularly acute in dexterous manipulation, e.g., teleoperating a multi-fingered hand for even a single atomic task can take days. To address this, we introduce Dexterous Point Policy, a framework that learns dexterous manipulation policies directly from human videos and requires no robot demonstrations. Our core insight is that a unified 3D keypoint representation can bridge human and robot embodiments when used for both observations and actions. Specifically, we extract 3D keypoints of task-relevant objects and human hands from raw videos, and train an autoregressive transformer over these keypoints. We observe that at the keypoint level, specifically the wrist and fingertips, human and robot behaviors closely align, enabling direct policy transfer. On a suite of real-robot tasks spanning pick-and-place and tool use, Dexterous Point Policy attains 75.0% success, whereas a state-of-the-art VLA baseline reaches only 1.0%. Furthermore, our method generalizes strongly to unseen scenarios, including multi-object environments and novel object categories.
