Similarity detection of three-dimensional object models

A system comprising:

• a memory with instructions stored thereon;
• and a processing device coupled to the memory, the processing device configured to execute the instructions to perform operations comprising:
• obtaining a three-dimensional (3D) model of an object, the 3D model including a mesh;
• executing a first neural network configured to use a diffusion operation to propagate feature information across a surface of the mesh;
• performing an area weighted averaging over features extracted from a plurality of vertices of the mesh to compute a mesh feature vector;
• computing a mesh distance score between the mesh feature vector and a reference encoded representation of a reference object;
• and
• determining that the object is a duplicate of the reference object if the mesh distance score meets a mesh threshold.

A non-transitory computer-readable medium with instructions stored thereon that, responsive to execution by a processing device, cause the processing device to perform operations comprising:

• obtaining a three-dimensional (3D) model including a mesh;
• processing the mesh using a first neural network that propagates feature information across a mesh surface using a diffusion operation;
• generating a mesh feature vector by performing an area weighted averaging over features of a plurality of vertices of the mesh surface;
• computing a mesh distance score between the mesh feature vector and a reference encoded representation of a reference object;
• and
• determining that the 3D model is a duplicate of the reference object if the mesh distance score meets a mesh threshold.