模型简介: ※经过训练的实验性LoRa 模型可以调整图片色彩饱和度。 这是一个通过100+张经过处理的图片训练的LoHa模型。 该模型可以提高背景的色彩饱和度、景深表现以及图像中轮廓之间连接的自然度。 权重的使用可以参考示例。 测试提示词和参数: prompt: sky, scenery, cloud, 1girl, building, outdoors, railing, solo, star (sky), city, long hair, cloudy sky, from below, blurry, starry sky, standing, wide shot, sunlight, skirt, dutch angle, brown hair, sitting, depth of field, stairs <lora:halation-noiseoffset-locon-32-000006:1> Negative prompt: fake animal ears,extra fingers,fewer fingers,(low quality, worst quality:1.65), (bad anatomy), (inaccurate limb:1.2),bad composition, inaccurate eyes, extra digit,fewer digits,(extra arms:1.2), Steps: 26, ENSD: 31337, Size: 448x768, Seed: 1557203104, Model: Counterfeit-V3.0_fp16, Sampler: DPM++ 2M Karras, CFG scale: 8, Clip skip: 2, Model hash: cbfba64e66, Hires steps: 20, Hires upscale: 2, Hires upscaler: ESRGAN_4x, Denoising strength: 0.45 |