EdgeFusion: On-Device Text-to-Image Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Castells, Thibault, Song, Hyoung-Kyu, Piao, Tairen, Choi, Shinkook, Kim, Bo-Kyeong, Yim, Hanyoung, Lee, Changgwun, Kim, Jae Gon, Kim, Tae-Ho
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929318661193728
author Castells, Thibault
Song, Hyoung-Kyu
Piao, Tairen
Choi, Shinkook
Kim, Bo-Kyeong
Yim, Hanyoung
Lee, Changgwun
Kim, Jae Gon
Kim, Tae-Ho
author_facet Castells, Thibault
Song, Hyoung-Kyu
Piao, Tairen
Choi, Shinkook
Kim, Bo-Kyeong
Yim, Hanyoung
Lee, Changgwun
Kim, Jae Gon
Kim, Tae-Ho
contents The intensive computational burden of Stable Diffusion (SD) for text-to-image generation poses a significant hurdle for its practical application. To tackle this challenge, recent research focuses on methods to reduce sampling steps, such as Latent Consistency Model (LCM), and on employing architectural optimizations, including pruning and knowledge distillation. Diverging from existing approaches, we uniquely start with a compact SD variant, BK-SDM. We observe that directly applying LCM to BK-SDM with commonly used crawled datasets yields unsatisfactory results. It leads us to develop two strategies: (1) leveraging high-quality image-text pairs from leading generative models and (2) designing an advanced distillation process tailored for LCM. Through our thorough exploration of quantization, profiling, and on-device deployment, we achieve rapid generation of photo-realistic, text-aligned images in just two steps, with latency under one second on resource-limited edge devices.
format Preprint
id arxiv_https___arxiv_org_abs_2404_11925
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle EdgeFusion: On-Device Text-to-Image Generation
Castells, Thibault
Song, Hyoung-Kyu
Piao, Tairen
Choi, Shinkook
Kim, Bo-Kyeong
Yim, Hanyoung
Lee, Changgwun
Kim, Jae Gon
Kim, Tae-Ho
Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
The intensive computational burden of Stable Diffusion (SD) for text-to-image generation poses a significant hurdle for its practical application. To tackle this challenge, recent research focuses on methods to reduce sampling steps, such as Latent Consistency Model (LCM), and on employing architectural optimizations, including pruning and knowledge distillation. Diverging from existing approaches, we uniquely start with a compact SD variant, BK-SDM. We observe that directly applying LCM to BK-SDM with commonly used crawled datasets yields unsatisfactory results. It leads us to develop two strategies: (1) leveraging high-quality image-text pairs from leading generative models and (2) designing an advanced distillation process tailored for LCM. Through our thorough exploration of quantization, profiling, and on-device deployment, we achieve rapid generation of photo-realistic, text-aligned images in just two steps, with latency under one second on resource-limited edge devices.
title EdgeFusion: On-Device Text-to-Image Generation
topic Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2404.11925