Text2Shape: Generating Shapes from Natural Language by Learning Joint Embeddings

TL;DR

This paper introduces a novel method that learns joint embeddings of natural language descriptions and 3D shapes, enabling effective text-to-shape retrieval and the generation of colored 3D objects from text descriptions, advancing cross-modal understanding.

Contribution

It presents the first approach to connect natural language with realistic 3D shapes using joint embeddings and a conditional Wasserstein GAN, enabling text-based shape generation.

Findings

Outperforms baseline methods in text-to-shape retrieval

Successfully generates colored 3D shapes from natural language

Creates a large dataset of language-shape pairs for training

Abstract

We present a method for generating colored 3D shapes from natural language. To this end, we first learn joint embeddings of freeform text descriptions and colored 3D shapes. Our model combines and extends learning by association and metric learning approaches to learn implicit cross-modal connections, and produces a joint representation that captures the many-to-many relations between language and physical properties of 3D shapes such as color and shape. To evaluate our approach, we collect a large dataset of natural language descriptions for physical 3D objects in the ShapeNet dataset. With this learned joint embedding we demonstrate text-to-shape retrieval that outperforms baseline approaches. Using our embeddings with a novel conditional Wasserstein GAN framework, we generate colored 3D shapes from text. Our method is the first to connect natural language text with realistic 3D objects exhibiting rich variations in color, texture, and shape detail. See video at https://youtu.be/zraPvRdl13Q