English

A 28.6 mJ/iter Stable Diffusion Processor for Text-to-Image Generation with Patch Similarity-based Sparsity Augmentation and Text-based Mixed-Precision

Hardware Architecture 2024-09-24 v3

Abstract

This paper presents an energy-efficient stable diffusion processor for text-to-image generation. While stable diffusion attained attention for high-quality image synthesis results, its inherent characteristics hinder its deployment on mobile platforms. The proposed processor achieves high throughput and energy efficiency with three key features as solutions: 1) Patch similarity-based sparsity augmentation (PSSA) to reduce external memory access (EMA) energy of self-attention score by 60.3 %, leading to 37.8 % total EMA energy reduction. 2) Text-based important pixel spotting (TIPS) to allow 44.8 % of the FFN layer workload to be processed with low-precision activation. 3) Dual-mode bit-slice core (DBSC) architecture to enhance energy efficiency in FFN layers by 43.0 %. The proposed processor is implemented in 28 nm CMOS technology and achieves 3.84 TOPS peak throughput with 225.6 mW average power consumption. In sum, 28.6 mJ/iteration highly energy-efficient text-to-image generation processor can be achieved at MS-COCO dataset.

Keywords

Cite

@article{arxiv.2403.04982,
  title  = {A 28.6 mJ/iter Stable Diffusion Processor for Text-to-Image Generation with Patch Similarity-based Sparsity Augmentation and Text-based Mixed-Precision},
  author = {Jiwon Choi and Wooyoung Jo and Seongyon Hong and Beomseok Kwon and Wonhoon Park and Hoi-Jun Yoo},
  journal= {arXiv preprint arXiv:2403.04982},
  year   = {2024}
}

Comments

Accepted at 2024 IEEE International Symposium on Circuits and Systems (ISCAS)

R2 v1 2026-06-28T15:13:03.918Z