Robotics

RoboSeg Maps Handles and Spouts From One Wrist Camera

Robb Harlan 4 min read

Grasping a watering-can handle is not the same problem as grasping its spout. Category labels and raw point clouds do not tell a robot which contact region matches a language task. Offline 3D part methods often assume a finished mesh first — which a tabletop arm does not have when it first sees the object.

RoboSeg, from Zhaochen Lan and Mengxiang Lin at Beihang University (arXiv:2608.09778, Aug 10, 2026), builds a part-labeled map online from a single eye-in-hand RGB-D camera, then indexes grasp proposals by those parts.

RoboSeg online part mapping versus offline mesh part segmentation
Offline mesh pipelines vs RoboSeg’s online eye-in-hand scan. Source: arXiv:2608.09778.

Pipeline

Hardware is a UR5 with an Intel RealSense D435i wrist camera. At start, Qwen3-VL looks at the first RGB frame and returns a short list of functional parts (for example, “watering can handle”). Scanning then runs two asynchronous threads:

  • Geometry: high-rate RGB-D odometry and TSDF fusion
  • Semantics: keyframe SAM3 masks for those part prompts, projected into voxels

Voxel-level temporal voting (top-3 instance IDs per voxel) dampens label flicker. After the scan, AnyGrasp proposes 6-DoF candidates on the reconstructed cloud; each candidate is tagged with the dominant part under the gripper close region, and the task language filters to the requested part.

RoboSeg three-module architecture diagram
VLM part discovery, asynchronous reconstruction, task-conditioned grasp selection. Source: arXiv:2608.09778.

Numbers

On manually labeled objects (Big Pot, Double Handle Pot, Watering Can, Kitchen Knife, Hammer), RoboSeg reports 83.4% mean part IoU. In a 24-trial physical pilot over eight object–task conditions (three repeats each), the executed grasp contacted the requested part in 24/24 trials. Combined physical lift + correct part succeeded in 21/24 trials; the three misses still touched the right part but failed the stable lift — pointing at grasp mechanics, not semantic routing. Unfiltered AnyGrasp top-1 matched the target part in only 4/8 conditions on recorded proposals, while semantic filtering selected a target-part candidate in 8/8.

Part-level qualitative comparison across pots, knife, and hammer
Functional parts on household tools: online RoboSeg vs offline baselines on clean geometry. Source: arXiv:2608.09778.

A Human’s Take

Semantic indexing on top of an off-the-shelf grasp detector is the right layering. You keep AnyGrasp for geometry and use the part map as a filter — that is how I want factory cells to grow features without rewriting the whole stack. Twenty-four trials is a pilot, not a production yield study, and thin blades still hurt physical success. If they open the code with the VLM prompt and voting config, this becomes a useful default for “grasp the handle, not the blade” demos.

Sources