Update README.md
Browse files
README.md
CHANGED
|
@@ -45,6 +45,7 @@ InstantCamera Triptych Results show off.
|
|
| 45 |
## Model Card for Model ID
|
| 46 |
|
| 47 |
因为原生的Qwen-Edit-2509的多图输入的结果中人物一致性并不是很好,特别是在一些指定的场景和指定的分辨率下。本模型重新针对多图输入的合照场景对Qwen-Edit-2509进行重新训练,保持高度人物一致性,输入两张图的图像为他们生成合照。
|
|
|
|
| 48 |
The native Qwen-Edit-2509 model often struggles with consistency in multi-image inputs, particularly at specific resolutions and scenes. This model is a fine-tuned version, specifically optimized for generating group photos from multiple portraits. It maintains high character consistency, allowing you to input individual photos and produce a coherent composite image.
|
| 49 |
|
| 50 |
|
|
@@ -55,10 +56,15 @@ The native Qwen-Edit-2509 model often struggles with consistency in multi-image
|
|
| 55 |
## Prompts phrase
|
| 56 |
|
| 57 |
```两人相互对视并牵手的照片,保持两人人脸、服饰、装扮高度一致,背景在温暖的客厅室内,95年代复古风格。整个画面具有老式照片的不清晰感,颗粒感,严重的晃动感,复古的滤镜。```
|
|
|
|
| 58 |
```两人相互接吻的照片,保持两人人脸、服饰、装扮高度一致,背景是平坦的高原山脉,80年代复古风格。整个画面具有老式照片的不清晰感,颗粒感,严重的晃动感,复古的滤镜。```
|
|
|
|
| 59 |
```两人相互对视并牵手的照片,保持两人人脸、服饰、装扮高度一致,背景在温暖的客厅室内,95年代复古风格。整个画面具有老式照片的不清晰感,颗粒感,严重的晃动感,复古的滤镜。```
|
|
|
|
| 60 |
```两人站在街头合照并挽着手的照片,保持两人人脸高度一致,男士穿着夹克外套,露出南瓜头的T恤,女士穿着时尚的偏橘黄色的装扮,背景是万圣节日的欧洲小镇的马路中央,万圣节节日风格```
|
|
|
|
| 61 |
```两人站在一个石头拱门墙前的酷酷合拍,保持两人人脸高度一致,两人穿着黑色的时尚帅气装扮,拱门后的天气阴暗,还有稀疏的枯木。```
|
|
|
|
| 62 |
```两人在室内的合照,保持两人人脸高度一致,两人穿着有万圣节节日气息的小饰品,背后有是奶油风的室内场景,有一些万圣节背景元素。```
|
| 63 |
|
| 64 |
There is no fixed trigger word. The specific photous prompt needs to be tested more.
|
|
@@ -67,7 +73,7 @@ There is no fixed trigger word. The specific photous prompt needs to be tested m
|
|
| 67 |
|
| 68 |
Weights for this model are available in Safetensors format.
|
| 69 |
|
| 70 |
-
[Download](https://huggingface.co/valiantcat/Qwen-Image-Edit-2509-photous)
|
| 71 |
|
| 72 |
## Training at Chongqing Valiant Cat
|
| 73 |
|
|
|
|
| 45 |
## Model Card for Model ID
|
| 46 |
|
| 47 |
因为原生的Qwen-Edit-2509的多图输入的结果中人物一致性并不是很好,特别是在一些指定的场景和指定的分辨率下。本模型重新针对多图输入的合照场景对Qwen-Edit-2509进行重新训练,保持高度人物一致性,输入两张图的图像为他们生成合照。
|
| 48 |
+
|
| 49 |
The native Qwen-Edit-2509 model often struggles with consistency in multi-image inputs, particularly at specific resolutions and scenes. This model is a fine-tuned version, specifically optimized for generating group photos from multiple portraits. It maintains high character consistency, allowing you to input individual photos and produce a coherent composite image.
|
| 50 |
|
| 51 |
|
|
|
|
| 56 |
## Prompts phrase
|
| 57 |
|
| 58 |
```两人相互对视并牵手的照片,保持两人人脸、服饰、装扮高度一致,背景在温暖的客厅室内,95年代复古风格。整个画面具有老式照片的不清晰感,颗粒感,严重的晃动感,复古的滤镜。```
|
| 59 |
+
|
| 60 |
```两人相互接吻的照片,保持两人人脸、服饰、装扮高度一致,背景是平坦的高原山脉,80年代复古风格。整个画面具有老式照片的不清晰感,颗粒感,严重的晃动感,复古的滤镜。```
|
| 61 |
+
|
| 62 |
```两人相互对视并牵手的照片,保持两人人脸、服饰、装扮高度一致,背景在温暖的客厅室内,95年代复古风格。整个画面具有老式照片的不清晰感,颗粒感,严重的晃动感,复古的滤镜。```
|
| 63 |
+
|
| 64 |
```两人站在街头合照并挽着手的照片,保持两人人脸高度一致,男士穿着夹克外套,露出南瓜头的T恤,女士穿着时尚的偏橘黄色的装扮,背景是万圣节日的欧洲小镇的马路中央,万圣节节日风格```
|
| 65 |
+
|
| 66 |
```两人站在一个石头拱门墙前的酷酷合拍,保持两人人脸高度一致,两人穿着黑色的时尚帅气装扮,拱门后的天气阴暗,还有稀疏的枯木。```
|
| 67 |
+
|
| 68 |
```两人在室内的合照,保持两人人脸高度一致,两人穿着有万圣节节日气息的小饰品,背后有是奶油风的室内场景,有一些万圣节背景元素。```
|
| 69 |
|
| 70 |
There is no fixed trigger word. The specific photous prompt needs to be tested more.
|
|
|
|
| 73 |
|
| 74 |
Weights for this model are available in Safetensors format.
|
| 75 |
|
| 76 |
+
[Download](https://huggingface.co/valiantcat/Qwen-Image-Edit-2509-photous/blob/main/QwenEdit2509_photous_000010000.safetensors)
|
| 77 |
|
| 78 |
## Training at Chongqing Valiant Cat
|
| 79 |
|