用视觉模型生成连续语义先验，让机器人识别未知物体姿态。
SCOPE: Semantic Conditioning for Sim2Real Category-Level Object Pose Estimation in Robotics
- 用DINOv2特征做连续语义先验，替代传统类别标签。
- 在合成数据上提升31.9%精度，未见物体抓取成功率最高达100%。
- 适合需要泛化到未知物体的机器人操作场景。
机器人操作依赖精准的物体位姿估计。在开放环境中，机器人会遇到未知物体，需具备语义理解能力以泛化至已知类别及之外。为此，我们提出SCOPE，一种基于扩散模型的类别级物体位姿估计方法，通过DINOv2特征作为连续语义先验，消除对离散类别标签的依赖。结合真实感训练数据与点法向量噪声模型，显著缩小了仿真到现实（Sim2Real）差距。通过交叉注意力注入连续语义先验，使模型学习跨实例的标准化坐标系，超越已知类别的分布。在合成训练下，5°5cm指标相对当前最优提升31.9%。在两个实例级数据集上的额外实验表明，其可泛化至未知类别，实现未见物体抓取成功率高达100%。代码已开源：https://github.com/hoenigpeter/scope。
原文摘要 · Abstract (English)
Object manipulation requires accurate object pose estimation. In open environments, robots encounter unknown objects, which requires semantic understanding in order to generalize both to known categories and beyond. To resolve this challenge, we present SCOPE, a diffusion-based category-level object pose estimation model that eliminates the need for discrete category labels by leveraging DINOv2 features as continuous semantic priors. By combining these DINOv2 features with photorealistic training data and a noise model for point normals, we reduce the Sim2Real gap in category-level object pose estimation. Furthermore, injecting the continuous semantic priors via cross-attention enables SCOPE to learn canonicalized object coordinate systems across object instances beyond the distribution of known categories. SCOPE outperforms the current state of the art in synthetically trained category-level object pose estimation, achieving a relative improvement of 31.9\% on the 5$^\circ$5cm metric. Additional experiments on two instance-level datasets demonstrate generalization beyond known object categories, enabling grasping of unseen objects from unknown categories with a success rate of up to 100\%. Code available: https://github.com/hoenigpeter/scope.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。