Shape from Shading for Robotic Manipulation
This work addresses shape estimation for robotic manipulation tasks, but it is incremental as it builds on existing shape-from-shading techniques with a specific workspace-scaled approach.
The authors tackled the problem of estimating object shape for robotic manipulation by using controlled illumination to capture surface normals and depth discontinuities, resulting in a method that supports grasping, deformation measurement, and pose estimation for table-top objects.
Controlling illumination can generate high quality information about object surface normals and depth discontinuities at a low computational cost. In this work we demonstrate a robot workspace-scaled controlled illumination approach that generates high quality information for table top scale objects for robotic manipulation. With our low angle of incidence directional illumination approach, we can precisely capture surface normals and depth discontinuities of monochromatic Lambertian objects. We show that this approach to shape estimation is 1) valuable for general purpose grasping with a single point vacuum gripper, 2) can measure the deformation of known objects, and 3) can estimate pose of known objects and track unknown objects in the robot's workspace.