If you find my model helpful, please light up the free collectible heart in the upper right corner, appreciate it.如果你觉得我的模型有帮助，请点亮右上角那个免费的收藏红心，不胜感激。<h3 id="heading-111">both:</h3><ul><li>Key feature: Enhanced eye detail throughout the face, rather than just a single eye close-up</li><li>主要特征：在整张面部进行眼睛细节的提升，而非仅仅是单只眼睛的特写</li><li>It usually makes the face more beautiful and adds details to the skin</li><li>通常会使人物面部更加漂亮，并增加皮肤细节</li><li>Can be mixed with other Loras for better facial results</li><li>可以与其他LoRA混合使用，获得更好的面部效果</li><li>Suggested weight: Whatever you want, just don't make it too big</li><li>建议权重：任何你想要的，别太大就好</li></ul><h3 id="heading-112">v2:</h3>Updated the training set and improved the resolution. However, facial collapse may occur when enlarging the full image by four times. It is recommended to enlarge it by only two times or use it for inpainting.<h3 id="heading-113">v1:</h3>More detailed eyes, which can be used for inpainting or directly generating pictures to improve the details of the eyes, as shown in the cover picture. Highres or img2img are recommended. To make the effect obvious.Because it is mainly trained on European and American faces, it will slightly affect the face and prefer European and American styles, so it can change the model of Asian faces.Recommended extensions:<ul><li><a target="_blank" rel="ugc" href="https://github.com/pkuliyi2015/multidiffusion-upscaler-for-automatic1111">https://github.com/pkuliyi2015/multidiffusion-upscaler-for-automatic1111</a></li><li><a target="_blank" rel="ugc" href="https://github.com/ototadana/sd-face-editor">https://github.com/ototadana/sd-face-editor</a></li></ul><h3 id="heading-114">v2:</h3>更新了训练集，提高了分辨率，细节更加丰富真实，但按照全图放大四倍的时候会出现面部崩坏，建议仅放大2倍或者用于inpainting。<h3 id="heading-115">v1:</h3>细节更加丰富的眼睛，可以用于inpainting，也可以直接生成图片以改善眼睛的细节，效果如封面图所示。建议使用Highres或者img2img放大。以使得效果明显。因为主要在欧美面孔上训练，故而会略微影响面部，偏向于欧美风格，因此可以变换亚洲面孔的模型。推荐的扩展：<ul><li><a target="_blank" rel="ugc" href="https://github.com/pkuliyi2015/multidiffusion-upscaler-for-automatic1111">https://github.com/pkuliyi2015/multidiffusion-upscaler-for-automatic1111</a></li><li><a target="_blank" rel="ugc" href="https://github.com/ototadana/sd-face-editor">https://github.com/ototadana/sd-face-editor</a></li></ul>Support me:<a target="_blank" rel="ugc" href="https://ko-fi.com/7dragons">https://ko-fi.com/7dragons</a>for china:<a target="_blank" rel="ugc" href="https://afdian.net/a/7dragonsAI">https://afdian.net/a/7dragonsAI</a>Thank you very much! Your help makes me more motivated to create!非常感谢！ 您的帮助让我更有动力创作！

Better eyes+face+skin | 更好的眼睛+脸+皮肤

v2:Updated the training set and improved the resolution. However, facial collapse may occur when enlarging the full image by four times. It is recommended to enlarge it by only two times or use it for inpainting.

Files with SHA256: 8b59bbea260189d5d177c9a15cb995cea09ff230cebe2a59fe47284dcda1f5e5