Latent Representation Matters: Human-like Sketches in One-shot Drawing Tasks

TL;DR

This paper investigates how different inductive biases in Latent Diffusion Models influence their ability to generate human-like sketches in one-shot drawing tasks, highlighting the importance of regularizations like redundancy reduction and prototypes.

Contribution

It systematically studies the impact of various inductive biases on LDMs' latent space, demonstrating that specific regularizations produce near-human-like sketches in one-shot tasks.

Findings

Redundancy reduction and prototype regularizations improve sketch recognizability.

LDMs with these regularizations produce sketches comparable to human perception.

The gap between human and machine one-shot drawing performance is nearly closed.

Abstract

Humans can effortlessly draw new categories from a single exemplar, a feat that has long posed a challenge for generative models. However, this gap has started to close with recent advances in diffusion models. This one-shot drawing task requires powerful inductive biases that have not been systematically investigated. Here, we study how different inductive biases shape the latent space of Latent Diffusion Models (LDMs). Along with standard LDM regularizers (KL and vector quantization), we explore supervised regularizations (including classification and prototype-based representation) and contrastive inductive biases (using SimCLR and redundancy reduction objectives). We demonstrate that LDMs with redundancy reduction and prototype-based regularizations produce near-human-like drawings (regarding both samples' recognizability and originality) -- better mimicking human perception (as evaluated psychophysically). Overall, our results suggest that the gap between humans and machines in one-shot drawings is almost closed.