ImageFolder: Autoregressive Image Generation with Folded Tokens

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Li, Xiang, Qiu, Kai, Chen, Hao, Kuen, Jason, Gu, Jiuxiang, Raj, Bhiksha, Lin, Zhe
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910726171394048
author Li, Xiang
Qiu, Kai
Chen, Hao
Kuen, Jason
Gu, Jiuxiang
Raj, Bhiksha
Lin, Zhe
author_facet Li, Xiang
Qiu, Kai
Chen, Hao
Kuen, Jason
Gu, Jiuxiang
Raj, Bhiksha
Lin, Zhe
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2410_01756
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle ImageFolder: Autoregressive Image Generation with Folded Tokens
Li, Xiang
Qiu, Kai
Chen, Hao
Kuen, Jason
Gu, Jiuxiang
Raj, Bhiksha
Lin, Zhe
Computer Vision and Pattern Recognition
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.
title ImageFolder: Autoregressive Image Generation with Folded Tokens
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2410.01756