Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Unleashing Transformers: Parallel Token Prediction with Discrete Absorbing Diffusion for Fast High-Resolution Image Generation from Vector-Quantized Codes

About

Whilst diffusion probabilistic models can generate high quality image content, key limitations remain in terms of both generating high-resolution imagery and their associated high computational requirements. Recent Vector-Quantized image models have overcome this limitation of image resolution but are prohibitively slow and unidirectional as they generate tokens via element-wise autoregressive sampling from the prior. By contrast, in this paper we propose a novel discrete diffusion probabilistic model prior which enables parallel prediction of Vector-Quantized tokens by using an unconstrained Transformer architecture as the backbone. During training, tokens are randomly masked in an order-agnostic manner and the Transformer learns to predict the original tokens. This parallelism of Vector-Quantized token prediction in turn facilitates unconditional generation of globally consistent high-resolution and diverse imagery at a fraction of the computational expense. In this manner, we can generate image resolutions exceeding that of the original training set samples whilst additionally provisioning per-image likelihood estimates (in a departure from generative adversarial approaches). Our approach achieves state-of-the-art results in terms of Density (LSUN Bedroom: 1.51; LSUN Churches: 1.12; FFHQ: 1.20) and Coverage (LSUN Bedroom: 0.83; LSUN Churches: 0.73; FFHQ: 0.80), and performs competitively on FID (LSUN Bedroom: 3.64; LSUN Churches: 4.07; FFHQ: 6.11) whilst offering advantages in terms of both computation and reduced training set requirements.

Sam Bond-Taylor, Peter Hessey, Hiroshi Sasaki, Toby P. Breckon, Chris G. Willcocks• 2021

Related benchmarks

TaskDatasetResultRank
Image GenerationLSUN Churches 256x256
FID4.07
23
Image GenerationFFHQ 1024x1024 (train)
FID6.11
23
Image GenerationLSUN Bedroom 256x256
FID3.64
16
Image GenerationLSUN Church 256x256 (train)
FID4.07
16
Image GenerationFFHQ 256x256 50k (test)
FID6.11
15
Image GenerationLSUN Church 256x256 50k (test)
FID4.07
10
Image GenerationFFHQ-256
FID6.11
8
Image GenerationFFHQ (test val)
Recall0.24
8
Image GenerationLSUN bedroom
Recall41
7
Image GenerationFFHQ 256x256
Precision73
5
Showing 10 of 10 rows

Other info

Code

Follow for update