A Neural Space-Time Representation for Text-to-Image Personalization

TL;DR

This paper introduces a novel space-time neural representation for text-to-image personalization, enabling high-quality, controllable image generation with a compact model that does not require fine-tuning the generative model.

Contribution

It proposes a new implicit space-time concept representation using a neural mapper, improving personalization quality and control without fine-tuning the generative model.

Findings

Enhanced visual fidelity and convergence in personalized image generation.

Ability to control concept reconstruction and editability with a single model.

Outperforms existing methods in generating high-quality, personalized images.

Abstract

A key aspect of text-to-image personalization methods is the manner in which the target concept is represented within the generative process. This choice greatly affects the visual fidelity, downstream editability, and disk space needed to store the learned concept. In this paper, we explore a new text-conditioning space that is dependent on both the denoising process timestep (time) and the denoising U-Net layers (space) and showcase its compelling properties. A single concept in the space-time representation is composed of hundreds of vectors, one for each combination of time and space, making this space challenging to optimize directly. Instead, we propose to implicitly represent a concept in this space by optimizing a small neural mapper that receives the current time and space parameters and outputs the matching token embedding. In doing so, the entire personalized concept is represented by the parameters of the learned mapper, resulting in a compact, yet expressive, representation. Similarly to other personalization methods, the output of our neural mapper resides in the input space of the text encoder. We observe that one can significantly improve the convergence and visual fidelity of the concept by introducing a textual bypass, where our neural mapper additionally outputs a residual that is added to the output of the text encoder. Finally, we show how one can impose an importance-based ordering over our implicit representation, providing users control over the reconstruction and editability of the learned concept using a single trained model. We demonstrate the effectiveness of our approach over a range of concepts and prompts, showing our method's ability to generate high-quality and controllable compositions without fine-tuning any parameters of the generative model itself.