ImageFolder: Autoregressive Image Generation with Folded Tokens

TL;DR

This paper introduces ImageFolder, a semantic tokenizer that uses folded tokens and dual-branch product quantization to improve autoregressive image generation quality and efficiency without increasing token length.

Contribution

The paper presents a novel semantic tokenizer with folded tokens and dual-branch quantization, balancing reconstruction and generation quality in autoregressive models.

Findings

ImageFolder achieves superior image generation quality.

It enables shorter token lengths without sacrificing performance.

The method improves efficiency in autoregressive image modeling.

Abstract

Image tokenizers are crucial for visual generative models, e.g., diffusion models (DMs) and autoregressive (AR) models, as they construct the latent representation for modeling. Increasing token length is a common approach to improve the image reconstruction quality. However, tokenizers with longer token lengths are not guaranteed to achieve better generation quality. There exists a trade-off between reconstruction and generation quality regarding token length. In this paper, we investigate the impact of token length on both image reconstruction and generation and provide a flexible solution to the tradeoff. We propose ImageFolder, a semantic tokenizer that provides spatially aligned image tokens that can be folded during autoregressive modeling to improve both generation efficiency and quality. To enhance the representative capability without increasing token length, we leverage dual-branch product quantization to capture different contexts of images. Specifically, semantic regularization is introduced in one branch to encourage compacted semantic information while another branch is designed to capture the remaining pixel-level details. Extensive experiments demonstrate the superior quality of image generation and shorter token length with ImageFolder tokenizer.