Binarized Diffusion Model for Image Super-Resolution

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Zheng, Qin, Haotong, Guo, Yong, Su, Xiongfei, Yuan, Xin, Kong, Linghe, Zhang, Yulun
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929569055899648
author Chen, Zheng
Qin, Haotong
Guo, Yong
Su, Xiongfei
Yuan, Xin
Kong, Linghe
Zhang, Yulun
author_facet Chen, Zheng
Qin, Haotong
Guo, Yong
Su, Xiongfei
Yuan, Xin
Kong, Linghe
Zhang, Yulun
contents Advanced diffusion models (DMs) perform impressively in image super-resolution (SR), but the high memory and computational costs hinder their deployment. Binarization, an ultra-compression algorithm, offers the potential for effectively accelerating DMs. Nonetheless, due to the model structure and the multi-step iterative attribute of DMs, existing binarization methods result in significant performance degradation. In this paper, we introduce a novel binarized diffusion model, BI-DiffSR, for image SR. First, for the model structure, we design a UNet architecture optimized for binarization. We propose the consistent-pixel-downsample (CP-Down) and consistent-pixel-upsample (CP-Up) to maintain dimension consistent and facilitate the full-precision information transfer. Meanwhile, we design the channel-shuffle-fusion (CS-Fusion) to enhance feature fusion in skip connection. Second, for the activation difference across timestep, we design the timestep-aware redistribution (TaR) and activation function (TaA). The TaR and TaA dynamically adjust the distribution of activations based on different timesteps, improving the flexibility and representation alability of the binarized module. Comprehensive experiments demonstrate that our BI-DiffSR outperforms existing binarization methods. Code is released at: https://github.com/zhengchen1999/BI-DiffSR.
format Preprint
id arxiv_https___arxiv_org_abs_2406_05723
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Binarized Diffusion Model for Image Super-Resolution
Chen, Zheng
Qin, Haotong
Guo, Yong
Su, Xiongfei
Yuan, Xin
Kong, Linghe
Zhang, Yulun
Computer Vision and Pattern Recognition
Advanced diffusion models (DMs) perform impressively in image super-resolution (SR), but the high memory and computational costs hinder their deployment. Binarization, an ultra-compression algorithm, offers the potential for effectively accelerating DMs. Nonetheless, due to the model structure and the multi-step iterative attribute of DMs, existing binarization methods result in significant performance degradation. In this paper, we introduce a novel binarized diffusion model, BI-DiffSR, for image SR. First, for the model structure, we design a UNet architecture optimized for binarization. We propose the consistent-pixel-downsample (CP-Down) and consistent-pixel-upsample (CP-Up) to maintain dimension consistent and facilitate the full-precision information transfer. Meanwhile, we design the channel-shuffle-fusion (CS-Fusion) to enhance feature fusion in skip connection. Second, for the activation difference across timestep, we design the timestep-aware redistribution (TaR) and activation function (TaA). The TaR and TaA dynamically adjust the distribution of activations based on different timesteps, improving the flexibility and representation alability of the binarized module. Comprehensive experiments demonstrate that our BI-DiffSR outperforms existing binarization methods. Code is released at: https://github.com/zhengchen1999/BI-DiffSR.
title Binarized Diffusion Model for Image Super-Resolution
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2406.05723