ScrewSplat: An End-to-End Method for Articulated Object Recognition
This addresses the problem of enabling robots to interact with everyday articulated objects like doors and laptops, representing a novel method for a known bottleneck.
The paper tackles articulated object recognition from RGB images by introducing ScrewSplat, an end-to-end method that simultaneously reconstructs 3D geometry and segments objects into movable parts. The method achieves state-of-the-art recognition accuracy and enables zero-shot, text-guided manipulation using the recovered kinematic model.
Articulated object recognition -- the task of identifying both the geometry and kinematic joints of objects with movable parts -- is essential for enabling robots to interact with everyday objects such as doors and laptops. However, existing approaches often rely on strong assumptions, such as a known number of articulated parts; require additional inputs, such as depth images; or involve complex intermediate steps that can introduce potential errors -- limiting their practicality in real-world settings. In this paper, we introduce ScrewSplat, a simple end-to-end method that operates solely on RGB observations. Our approach begins by randomly initializing screw axes, which are then iteratively optimized to recover the object's underlying kinematic structure. By integrating with Gaussian Splatting, we simultaneously reconstruct the 3D geometry and segment the object into rigid, movable parts. We demonstrate that our method achieves state-of-the-art recognition accuracy across a diverse set of articulated objects, and further enables zero-shot, text-guided manipulation using the recovered kinematic model. See the project website at: https://screwsplat.github.io.