PoseVLA: 汎用ポーズ事前学習による汎化可能な視覚-言語-行動ポリシー
PoseVLA: Universal Pose Pretraining for Generalizable Vision-Language-Action Policies
VLAモデルの特徴崩壊と学習効率の低さを解決するため、カメラ中心の統一空間で3D空間事前知識を抽出する事前学習と、ロボット固有の行動空間への適応を行う事後学習に分離したPose-VLAを提案。離散ポーズトークンにより多様な3Dデータと軌道データを統合し、RoboTwin 2.0で79.5%、LIBEROで96.0%の成功率を達成。
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: Haitao Lin, Hanyang Yu, Jingshun Huang, He Zhang, Yonggen Ling, Ping Tan, Xiangyang Xue, Yanwei Fu
分類: cs.CV, cs.LG, cs.RO
原文アブストラクト
Existing Vision-Language-Action (VLA) models often suffer from feature collapse and low training efficiency because they entangle high-level perception with sparse, embodiment-specific action supervision. Since these models typically rely on VLM backbones optimized for Visual Question Answering (VQA), they excel at semantic identification but often overlook subtle 3D state variations that dictate distinct action patterns. To resolve these misalignments, we propose Pose-VLA, a decoupled paradigm that separates VLA training into a pre-training phase for extracting universal 3D spatial priors in a unified camera-centric space, and a post-training phase for efficient embodiment alignment within robot-specific action space. By introducing discrete pose tokens as a universal representation, Pose-VLA seamlessly integrates spatial grounding from diverse 3D datasets with geometry-level trajectories from robotic demonstrations. Our framework follows a two-stage pre-training pipeline, establishing fundamental spatial grounding via poses followed by motion alignment through trajectory supervision. Extensive evaluations demonstrate that Pose-VLA achieves state-of-the-art results on RoboTwin 2.0 with a 79.5% average success rate and competitive performance on LIBERO at 96.0%. Real-world experiments further showcase robust generalization across diverse objects using only 100 demonstrations per task, validating the efficiency of our pre-training paradigm.
