Weixiao Liu

@qk98 Hi, thanks for your interest in our algorithm. This might happen, but could be mitigated by the following tips: First, allow more iterations for each superquadric optimization (maxOptiIter), which...

Currently, the mesh is generated by a very simple algorithm (which can be found in the source code) just for shape visualization purpose, without face orientation computed. The meshlab guesses...

> Get it. But I didn't find an option in Meshlab to "render the mesh without normals". How did you visualize the meshes in your paper? We used open3d or...

> Hi, is there any progress on the Python implementation? Working on it. I just defended my dissertation recently, and kind of busy. Sorry for the delay.

Thanks for your interest. The examples contain comparisons between 3D reconstruction/abstraction from marching cubes and marching primitives from the same SDF. *_mc.stl is obtained from marching cubes, and *_mps.stl is...

The superquadrics are not saved in parametric form. The files .stl are triangle meshes which are geometry approximations of the superquadrics for visualization on computers only. This is because computer...

> Hi, > > Thanks for your excellent work! I'm currently working towards recovery of quadrics from range data, but when I tried to implement the normal superquadrics fitting algorithm...