Shifted Diffusion for Text-to-image Generation

被引:15
|
作者
Zhou, Yufan [1 ]
Liu, Bingchen [2 ]
Zhu, Yizhe [2 ]
Yang, Xiao [2 ]
Chen, Changyou [1 ]
Xu, Jinhui [1 ]
机构
[1] SUNY Buffalo, Buffalo, NY 14260 USA
[2] ByteDance, Beijing, Peoples R China
关键词
D O I
10.1109/CVPR52729.2023.00979
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We present Corgi, a novel method for text-to-image generation. Corgi is based on our proposed shifted diffusion model, which achieves better image embedding generation from input text. Unlike the baseline diffusion model used in DALL-E 2, our method seamlessly encodes prior knowledge of the pre-trained CLIP model in its diffusion process by designing a new initialization distribution and a new transition step of the diffusion. Compared to the strong DALL-E 2 baseline, our method performs better in generating image embedding from the text in terms of both efficiency and effectiveness, resulting in better text-to-image generation. Extensive large-scale experiments are conducted and evaluated in terms of both quantitative measures and human evaluation, indicating a stronger generation ability of our method compared to existing ones. Furthermore, our model enables semi-supervised and language-free training for text-to-image generation, where only part or none of the images in the training dataset have an associated caption. Trained with only 1.7% of the images being captioned, our semi-supervised model obtains FID results comparable to DALL-E 2 on zero-shot text-to-image generation evaluated on MS-COCO. Corgi also achieves new state-of-the-art results across different datasets on downstream language-free text-to-image generation tasks, outperforming the previous method, Lafite, by a large margin.
引用
收藏
页码:10157 / 10166
页数:10
相关论文
共 50 条
  • [1] EmoGen: Emotional Image Content Generation with Text-to-Image Diffusion Models
    Yang, Jingyuan
    Feng, Jiawei
    Huang, Hui
    2024 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2024, 2024, : 6358 - 6368
  • [2] Controllable Text-to-Image Generation
    Li, Bowen
    Qi, Xiaojuan
    Lukasiewicz, Thomas
    Torr, Philip H. S.
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 32 (NIPS 2019), 2019, 32
  • [3] Surgical text-to-image generation
    Nwoye, Chinedu Innocent
    Bose, Rupak
    Elgohary, Kareem
    Arboit, Lorenzo
    Carlino, Giorgio
    Lavanchy, Joel L.
    Mascagni, Pietro
    Padoy, Nicolas
    PATTERN RECOGNITION LETTERS, 2025, 190 : 73 - 80
  • [4] Expressive Text-to-Image Generation with Rich Text
    Ge, Songwei
    Park, Taesung
    Zhu, Jun-Yan
    Huang, Jia-Bin
    2023 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION, ICCV, 2023, : 7511 - 7522
  • [5] MagicFusion: Boosting Text-to-Image Generation Performance by Fusing Diffusion Models
    Zhao, Jing
    Zheng, Heliang
    Wang, Chaoyue
    Lan, Long
    Yang, Wenjing
    2023 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2023), 2023, : 22535 - 22545
  • [6] Comparative Review of Text-to-Image Generation Techniques Based on Diffusion Models
    Gao, Xinyu
    Du, Fang
    Song, Lijuan
    Computer Engineering and Applications, 2024, 60 (24) : 44 - 64
  • [7] Debiasing Text-to-Image Diffusion Models
    He, Ruifei
    Xue, Chuhui
    Tan, Haoru
    Zhang, Wenqing
    Yu, Yingchen
    Bai, Song
    Qi, Xiaojuan
    PROCEEDINGS OF THE 1ST ACM MULTIMEDIA WORKSHOP ON MULTI-MODAL MISINFORMATION GOVERNANCE IN THE ERA OF FOUNDATION MODELS, MIS 2024, 2024, : 29 - 36
  • [8] RAPHAEL: Text-to-Image Generation via Large Mixture of Diffusion Paths
    Xue, Zeyue
    Song, Guanglu
    Guo, Qiushan
    Liu, Boxiao
    Zong, Zhuofan
    Liu, Yu
    Luo, Ping
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 36 (NEURIPS 2023), 2023,
  • [9] SEMANTICALLY INVARIANT TEXT-TO-IMAGE GENERATION
    Sah, Shagan
    Peri, Dheeraj
    Shringi, Ameya
    Zhang, Chi
    Dominguez, Miguel
    Savakis, Andreas
    Ptucha, Ray
    2018 25TH IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), 2018, : 3783 - 3787
  • [10] Text-to-Image Generation for Abstract Concepts
    Liao, Jiayi
    Chen, Xu
    Fu, Qiang
    Du, Lun
    He, Xiangnan
    Wang, Xiang
    Han, Shi
    Zhang, Dongmei
    THIRTY-EIGHTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, VOL 38 NO 4, 2024, : 3360 - 3368