首頁Home › 人工智慧與新興科技 AI & Emerging Tech
AI繪圖
1這是什麼What is it?
AI 繪圖指以生成模型依文字或圖像條件產生圖像的技術與實踐,2022 年 DALL·E 2、Midjourney 與 Stable Diffusion 陸續公開後迅速普及。其流程涉及提示詞、取樣參數、種子值與後續的修補(inpainting)、放大與控制(如 ControlNet);創作的作者性、風格模仿與訓練資料來源,是美學與法律上的核心爭議。
AI drawing refers to techniques and practices of producing images from text or image conditions with generative models, spreading rapidly after DALL·E 2, Midjourney and Stable Diffusion became public in 2022. Its workflow involves prompts, sampling parameters, seed values and subsequent inpainting, upscaling and control (such as ControlNet), and authorship, style imitation and training-data provenance are core aesthetic and legal controversies.
2常被用在哪種研究問題Typical research questions
常見問法:AI 生成圖像與人工圖像在評價上有何差異?
AI 繪圖的品質與風格如何被評價,值得探討。本研究以觀者問卷比較 AI 與人類作品的美感與真實感評分。
How the quality and style of AI-generated images are evaluated merits study. This study compares viewers' aesthetic and authenticity ratings of AI and human works by questionnaire.
撰寫內容,非真實研究結果,請與指導教授討論後改寫。
Written as a model example, not a real finding. Please revise with your advisor.
3怎麼測它How to measure it
盲評實驗+評分、辨識測試
Measurement notes are written in Chinese. Switch to 中/EN to read them.
4代表文獻Key references
英文論文(2000 年後)English paper (post-2000)
Mazzone, M., & Elgammal, A. (2019). Art, creativity, and the potential of artificial intelligence. Arts, 8(1).
中文論文(2000 年後)
中文論文為預填檢索連結,請自行挑選一篇 2000 年後的期刊論文並核對作者、年份、出處後再引用。
英文引用來自模型整理,引用前請務必核對並閱讀原文。English citations are compiled by a model. Verify and read the original before citing.