Spaces:
Running
Running
| import gradio as gr | |
| from transformers import pipeline | |
| # 建立 image-classification pipeline,使用 FaceAIorNot 模型 | |
| pipe = pipeline( | |
| task="image-classification", | |
| model="hchcsuim/FaceAIorNot" | |
| ) | |
| # 預測函數:回傳圖片和分類結果 | |
| def predict(input_img): | |
| predictions = pipe(input_img) | |
| return input_img, {p["label"]: p["score"] for p in predictions} | |
| # 精簡中英文對照版說明 | |
| model_card = """ | |
| --- | |
| ## 🧠 Model Description / 模型簡介 | |
| This model classifies face images into two categories: **AI-generated** and **Not AI-generated**. | |
| 本模型將臉部圖片分類為「AI生成」或「非AI生成」。 | |
| 🔗 Model Homepage / 模型主頁:https://huggingface.co/hchcsuim/FaceAIorNot | |
| --- | |
| ### 📊 Evaluation Results / 模型評估 | |
| - Accuracy 準確率: **99.35%** | |
| - Precision 精確率: **99.25%** | |
| - Recall 召回率: **99.47%** | |
| - F1-score F1 分數: **99.36%** | |
| - Loss 損失: **0.0233** | |
| Dataset: 105,330 face images (50% real / 50% AI), from 17 datasets, 14 generation techniques, 90% train / 10% test | |
| 資料集共 105,330 張臉部圖片(50% 真人 / 50% AI),來自 17 個資料集、14 種生成技術,90% 訓練 / 10% 測試 | |
| --- | |
| ### ⚠️ Disclaimer / 免責聲明 | |
| This model is for research and educational use only. Not 100% accurate. | |
| 請僅用於研究與教育用途,結果非百分之百準確。 | |
| Do not use for identity verification, legal judgment, or public exposure. | |
| 請勿用於身分驗證、法律判斷或公開揭露等用途。 | |
| --- | |
| ### 👨💻 About the Developer / 關於開發者 | |
| **Hung Chih Hsiang(洪誌翔)** | |
| Master’s student in Information Management, Cheng Shiu University. | |
| 正修科技大學 資訊管理所碩士生。 | |
| Focus: deepfake detection, snake recognition, system development, DevOps | |
| 專長:深度偽造辨識、蛇類辨識、系統開發與運維 | |
| --- | |
| ### 🤝 Open to Collaborations / 歡迎合作 | |
| I'm open to: | |
| 我歡迎以下方向的合作: | |
| - Research or commercial AI projects / AI 研究或商業應用 | |
| - Model training, optimization, deployment / 模型訓練、優化與部署 | |
| - System development, MLOps & DevOps / 系統開發與 MLOps、DevOps 整合 | |
| --- | |
| 📧 Email: hchcsuim@gmail.com | |
| 🐙 GitHub: https://github.com/hchcsuim | |
| """ | |
| # Gradio 介面設定(雙語版) | |
| gradio_app = gr.Interface( | |
| fn=predict, | |
| inputs=gr.Image(label="📸 Select / Upload Face Photo 選擇或上傳臉部照片", sources=["upload", "webcam"], type="pil"), | |
| outputs=[ | |
| gr.Image(label="🖼️ Input Image / 輸入圖片"), | |
| gr.Label(label="🔍 Classification Result / 判斷結果", num_top_classes=2) | |
| ], | |
| title="FaceAIorNot | 真人臉,還是 AI 生成人臉?", | |
| description=( | |
| "🤖 Upload or take a face photo to see if it's AI-generated or real.\n" | |
| "🧑 上傳或拍攝一張臉部照片,判斷是真人還是 AI 生成圖。" | |
| ), | |
| article=model_card, | |
| allow_flagging="never" | |
| ) | |
| if __name__ == "__main__": | |
| gradio_app.launch() | |