Compositional Conditional Diffusion Model

Shih Lun Lai, Pin Chuan Chen, Ching Wen Ma

研究成果: Conference contribution同行評審

摘要

Generative model often struggle to produce data beyond the training distribution. To address this, we explored various methods for unseen data generation, eventually focusing on compositional zero-shot image generation. By guiding the generation model with compositional class labels, we achieved better control over the generation process. Our model can generate unseen images whose compositional labels are not appear in the training set. While large language models like GPT-4 and image generation models like DALL-E offer similar zero-shot generation capabilities, our research emphasizes domain-specific zero-shot generation using smaller models. Through a series of tasks, we demonstrated the effectiveness of compositional zero-shot image generation across various complexities, showcasing its potential in contemporary machine learning.

原文English
主出版物標題Proceedings of the 2024 10th International Conference on Applied System Innovation, ICASI 2024
編輯Shoou-Jinn Chang, Sheng-Joue Young, Artde Donald Kin-Tak Lam, Liang-Wen Ji, Stephen D. Prior
發行者Institute of Electrical and Electronics Engineers Inc.
頁面377-379
頁數3
ISBN(電子)9798350394924
DOIs
出版狀態Published - 2024
事件10th International Conference on Applied System Innovation, ICASI 2024 - Kyoto, 日本
持續時間: 17 4月 202421 4月 2024

出版系列

名字Proceedings of the 2024 10th International Conference on Applied System Innovation, ICASI 2024

Conference

Conference10th International Conference on Applied System Innovation, ICASI 2024
國家/地區日本
城市Kyoto
期間17/04/2421/04/24

指紋

深入研究「Compositional Conditional Diffusion Model」主題。共同形成了獨特的指紋。

引用此