Canopus LoRA Flux FaceRealism

🚀 Canopus-LoRA-Flux-FaceRealism

Canopus-LoRA-Flux-FaceRealism 是一款圖像生成模型，藉助 LoRA 微調技術，專注於生成逼真的面部圖像，能在多種場景下輸出高質量的圖像效果。

🚀 快速開始

代碼示例

import torch
from pipelines import DiffusionPipeline

base_model = "black-forest-labs/FLUX.1-dev"
pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)

lora_repo = "prithivMLmods/Canopus-LoRA-Flux-FaceRealism"
trigger_word = "Realism"  # 若不使用觸發詞，可留空。
pipe.load_lora_weights(lora_repo)

device = torch.device("cuda")
pipe.to(device)

✨ 主要特性

• 高逼真度：能夠生成極其逼真的面部圖像，還原細膩的紋理和真實的光影效果。
• 多場景適用：可用於生成不同風格和場景的圖像，如時尚攝影、人像寫真等。
• 易於使用：通過簡單的觸發詞即可觸發圖像生成。

📦 安裝指南

暫未提及具體安裝步驟，可參考代碼示例中的導入和加載操作。

💻 使用示例

基礎用法

# 上述快速開始中的代碼示例即為基礎用法
import torch
from pipelines import DiffusionPipeline

base_model = "black-forest-labs/FLUX.1-dev"
pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)

lora_repo = "prithivMLmods/Canopus-LoRA-Flux-FaceRealism"
trigger_word = "Realism"  # 若不使用觸發詞，可留空。
pipe.load_lora_weights(lora_repo)

device = torch.device("cuda")
pipe.to(device)

高級用法

可根據不同的觸發提示詞生成不同風格的圖像，例如：

A portrait of an attractive woman in her late twenties with light brown hair and purple, wearing large a a yellow sweater. She is looking directly at the camera, standing outdoors near trees.. --ar 128:85 --v 6.0 --style raw

📚 詳細文檔

模型描述

prithivMLmods/Canopus-LoRA-Flux-FaceRealism

圖像處理參數

標註信息

標註：florence2 - en（自然語言 & 英語）

訓練數據

總共使用 200 + [高分辨率] 圖像進行訓練。

新版本信息

這裡有一個表格總結了 Hugging Face 上 Flux - Super - Realism - LoRA 模型的相關信息：

圖像對比 DLC

觸發提示

以下是一些觸發提示示例：

A portrait of an attractive woman in her late twenties with light brown hair and purple, wearing large a a yellow sweater. She is looking directly at the camera, standing outdoors near trees.. --ar 128:85 --v 6.0 --style raw

Headshot of handsome young man, wearing dark gray sweater with buttons and big shawl collar, brown hair and short beard, serious look on his face, black background, soft studio lighting, portrait photography --ar 85:128 --v 6.0 --style raw

A photo of the model wearing a white bodysuit and beige trench coat, posing in front of a train station with hands on head, soft light, sunset, fashion photography, high resolution, 35mm lens, f/22, natural lighting, global illumination. --ar 85:128 --v 6.0 --style raw

An image of a woman with fair skin and natural, wavy hair styled in soft curls around her face. She has defined eyebrows and light-colored eyes that draw attention. She's dressed in a red and navy plaid shirt with the top unbuttoned to show a white undershirt, and the sleeves rolled up to the forearms. The woman is casually leaning against a weathered blue door frame with peeling paint, adding a rustic charm to the scene. Her arms are crossed or resting in front of her, and she has a soft, contemplative expression on her face.

A black and white photo of a woman with shadow., in the style of minimalist beauty, light bronze and bronze, serene faces, sharp edges, karencore, high definition, contrast shading, realism --ar 3:5 --stylize 750 --v 6.0 --style raw

Photograph, face, young man professional model, minimalist, stylish nessy long hairstyle, beard, profound gaze, Canon EOS R5 with an RF 85mm F1.2L lens, solid white environment, studio lights setting, ultra-realistic, 32K UHD --ar 9:16 --style raw --stylize 700

觸發參數

觸發詞

你可以使用以下觸發詞來觸發圖像生成：

• face
• realism
• face realism

應用文件結構

/project-root/
├── .gitattributes
├── README.md
├── app.py
├── pythonproject.py

下載模型

該模型的權重以 Safetensors 格式提供。 點擊下載，在 Files & versions 標籤中獲取。

• 🪄 動漫模型：https://huggingface.co/prithivMLmods/Canopus-Anime-Art-Flux-LoRA

🤗 更多信息：https://hf.co/prithivmlmods

🔧 技術細節

暫未提供足夠詳細的技術實現細節。

📄 許可證

該模型使用 creativeml - openrail - m 許可證。

⚠️ 重要提示

請勿使用負提示詞，因為這可能會導致生成的圖像出現偽影和扭曲。