Concept Activation Vectors for Generating User-Defined 3D Shapes

TL;DR

This paper introduces a method using Concept Activation Vectors to interpret and manipulate 3D shape representations in deep learning models, enabling user-defined concept-based modifications in CAD designs.

Contribution

It presents a novel approach to interpret 3D shape embeddings with CAVs, allowing concept-based editing and analysis in geometric deep learning models.

Findings

CAVs effectively encode user-defined concepts in 3D shape space.

The method enables targeted modifications of 3D designs based on concepts.

Statistical tests confirm the significance of identified concepts.

Abstract

We explore the interpretability of 3D geometric deep learning models in the context of Computer-Aided Design (CAD). The field of parametric CAD can be limited by the difficulty of expressing high-level design concepts in terms of a few numeric parameters. In this paper, we use a deep learning architectures to encode high dimensional 3D shapes into a vectorized latent representation that can be used to describe arbitrary concepts. Specifically, we train a simple auto-encoder to parameterize a dataset of complex shapes. To understand the latent encoded space, we use the idea of Concept Activation Vectors (CAV) to reinterpret the latent space in terms of user-defined concepts. This allows modification of a reference design to exhibit more or fewer characteristics of a chosen concept or group of concepts. We also test the statistical significance of the identified concepts and determine the sensitivity of a physical quantity of interest across the dataset.