0

MaskGIT: Masked Generative Image Transformer

MaskGIT, a bidirectional transformer decoder, synthesizes high-fidelity images more efficiently than existing models by predicting all tokens simultaneously and refining iteratively.

Year
2022
Venue
CVPR 2022 1
Authors
5
Hosting
Abstract onlyARXIV-DEFAULT

Cite

Notes

Only stored in your browser.

Attribution

Abstract & full text
arxiv.org/abs/2202.04200ARXIV-DEFAULT
TL;DR
Semantic Scholar
Attribution policy →

Abstract

Generative transformers have experienced rapid popularity growth in the computer vision community in synthesizing high-fidelity and high-resolution images. The best generative transformer models so far, however, still treat an image naively as a sequence of tokens, and decode an image sequentially following the raster scan ordering (i.e. line-by-line). We find this strategy neither optimal nor efficient. This paper proposes a novel image synthesis paradigm using a bidirectional transformer decoder, which we term MaskGIT. During training, MaskGIT learns to predict randomly masked tokens by attending to tokens in all directions. At inference time, the model begins with generating all tokens of an image simultaneously, and then refines the image iteratively conditioned on the previous generation. Our experiments demonstrate that MaskGIT significantly outperforms the state-of-the-art transformer model on the ImageNet dataset, and accelerates autoregressive decoding by up to 64x. Besides, we illustrate that MaskGIT can be easily extended to various image editing tasks, such as inpainting, extrapolation, and image manipulation.

Authors

5