TCP4D: A Unified Tool-Center-Point Tracking and Generation Model for Robot Manipulation

Media reports Hugging Face Papers Source language: Chinese

The paper introduces TCP4D, a unified model that tracks tool-center-point poses and generates future trajectories from monocular video, achieving absolute trajectory errors of 7.60 mm and 16.07 mm on DROID and RoboLab, and reaching 69.19% mean success on RoboTwin, with joint training improving target-robot success by 41.5%.

Full text

Robot manipulation videos provide a rich visual experience, yet recovering robot action and gripper-state annotations from RGB observations still remains challenging. Using the embodiment-agnostic tool-center-point (TCP) pose as the action space, we present TCP4D, a unified model in which a shared 4D scene representation both tracks observed TCP motion and generates future TCP trajectories. Given a monocular video and initial 2D TCP queries, the tracker directly recovers metric TCP poses in SE(3) and gripper states alongside scene geometry and camera parameters. The generator samples local features from the same 4D representation along the recovered TCP tracks, then combines them with robot observations and language instructions to predict future TCP states. Our new cross-embodiment augmentation pipeline converts existing RGB videos into demonstrations for alternative embodiments by retargeting the recovered motion and synthesizing the corresponding observations, without recorded source-robot states. Experiments show that the tracker achieves absolute trajectory errors of 7.60 and 16.07 mm on DROID and RoboLab and adapts to a new embodiment with only ten demonstrations. The generator achieves 69.19% mean success on RoboTwin and outperforms well-established policy models on two real-world tasks, while joint training on source and augmented demonstrations improves target-robot success by 41.5% over augmented-only training.

View arXiv page View PDF Project page GitHub 7 Add to collection

Sources

← Back to Daily AI Report