场景:
这段时间成迷于AI绘画,目前已经学会了视频转成二次元风格。这里利用了秋葉aaaki大佬的Stable diffusion WebUI,和mov2mov插件,在controlnet 插件中利用了openpose来约束每张图的动作和手部,减少出现多只手和多个脚,同时还利用了controlnet插件中的depth来约束一下整体人物,减少出现多个人。整体效果我感觉还可以,只是感觉整个人往后仰,我还不确定这是mov2mov本身问题还是参数中出了问题(主要来一次太长时间了,19秒就要3090训练6h我一度怀疑是不是显卡已经拉掉了),还在调整中,,,,,,来看一下成片和效果。
生成视频图的参数设置
lora:shenheLoraCollection_shenheHard:0.4,((1woman)), ((solo)), (8k, RAW photo, best quality, masterpiece:1.2),(realistic, photo-realistic:1.37), detailed beautiful face,(8k, RAW photo, best quality, masterpiece:1.2), (photorealistic:1.4), (soft focus:1.4), (sharp focus:1.4), ultra-detailed,cute, solo,beautiful detailed sky,detailed cafe,(best quality), ((realistic photo)), (ultra detailed:1.3), (((cinematic color grading))), smile, (platium pink hair:1.2), warm sunlight, white teeth, lora:raidenShogunLora_raidenShogun:0.3,(large breasts), full body, (masterpiece, best quality, ultra-detailed, highres), (sidelighting, lustrous skin), (bloom), (shine), lighting, bokeh, ray tracing, long hair, lora:namiOnePiecePreAndPost_postTimeskip:0.4,
Negative prompt: :paintings, sketches, (worst quality:2), (low quality:2), (normal quality:2), lowres, normal quality, ((monochrome)), ((grayscale)), skin spots, acnes, skin blemishes, age spot, (outdoor:1.6), manboobs, backlight,(ugly:1.331), (duplicate:1.331), (morbid:1.21), (mutilated:1.21), (tranny:1.331), mutated hands, (poorly drawn hands:1.331), blurry, (bad anatomy:1.21), (bad proportions:1.331), extra limbs, (disfigured:1.331), (more than 2 nipples:1.331), (missing arms:1.331), (extra legs:1.331), (fused fingers:1.61051), (too many fingers:1.61051), (unclear eyes:1.331), bad hands, missing fingers, extra digit, (futa:1.1), bad body, NG_DeepNegative_V1_75T,pubic hair, glans
Steps: 40, Sampler: Euler a, CFG scale: 10, Seed: 1173863503, Size: 690x960, Model hash: 4078eb4174, Model: 3Guofeng3_v33, Denoising strength: 0.3, Clip skip: 2, ENSD: 31337, Mask blur: 4, ControlNet-0 Enabled: True, ControlNet-0 Module: openpose_hand, ControlNet-0 Model: control_openpose [b46e25f5], ControlNet-0 Weight: 0.35, ControlNet-0 Guidance Start: 0, ControlNet-0 Guidance End: 1, Noise multiplier: 0文章来源:https://www.toymoban.com/news/detail-504128.html
效果展示:
1681507739文章来源地址https://www.toymoban.com/news/detail-504128.html
大佬们懂的都懂,我怕真人有什么版权,肖像权问题,我就不放了,谨慎中,之前也是因为此类问题,错失了去某高校读博的机会(这是后来如同我再造父母的好心人告诉我的)。这个视频出现的抖音号也不是我的,我也不会视频删水印和,如果有问题,好和我说一下,我也学习一波。
到了这里,关于Stable diffusion中mov2mov的文章就介绍完了。如果您还想了解更多内容,请在右上角搜索TOY模板网以前的文章或继续浏览下面的相关文章,希望大家以后多多支持TOY模板网!