（批量）服装生成摄影级案例
5.0
0 reviews
Description
最近很火的换装换产品工作流，在其中加入了提词反推和自动遮罩实现了可以批量一键生成摄影级服装摄影图
工作流逻辑是通过上传的图片反推为提示词第一遍生成穿着相似衣服的模特，随后在通过fill/redux模型进行换装放大。
如果要上传图片进行换装就把模特图放入下面黄色节点内，同时关闭模特+flux生成
如果需要自行遮罩就把自动遮罩关闭。
如果最终放大效果不好就忽略
如果需要批量出图，将需要批量重绘的服装图片，放入一个文件夹内随后复制文件夹的路径填入加载批次图像的路径中，并将图像输出俩捏到右侧红色转接点，选择添加提示词队列下方的批次数量填为文件夹内需要重绘的图片数量。
该工作流对小白很友上传服装图下好模型依次运行就好
模型用到了flux-fill大模型：https://huggingface.co/black-forest-labs/FLUX.1-Fill-dev/tree/main
FLUX.1-Redux-dev：https://huggingface.co/black-forest-labs/FLUX.1-Fill-dev/tree/main
后续会分享更多好用的工作流，想解锁更多商用工作流可以+我Aiyo_pic
Discussion
(No comments yet)
Loading...
Reviews
No reviews yet
Versions (1)
- latest (a year ago)
Node Details
Primitive Nodes (25)
Any Switch (rgthree) (3)
Anything Everywhere3 (1)
BizyAirJoyCaption2 (1)
BizyAirSiliconCloudLLMAPI (1)
CLIPTextEncodeFlux (1)
Fast Groups Muter (rgthree) (1)
FluxGuidance (2)
Image Comparer (rgthree) (2)
ImageConcatMulti (1)
LayerMask: SegformerFashionPipelineLoader (1)
LayerMask: SegformerUltraV2 (2)
LongCLIPTextEncodeFlux (1)
Note (1)
Prompts Everywhere (1)
Reroute (1)
StyleModelApplySimple (1)
TTP_Image_Assy (1)
TTP_Image_Tile_Batch (1)
TTP_Tile_image_size (1)
easy imageBatchToImageList (1)
Custom Nodes (94)
- DeepTranslatorTextNode (1)
- CR Text Replace (3)
ComfyUI
- VAEDecode (2)
- VAELoader (1)
- UNETLoader (3)
- CLIPVisionLoader (1)
- CLIPVisionEncode (1)
- InpaintModelConditioning (1)
- ConditioningZeroOut (1)
- KSampler (1)
- DifferentialDiffusion (1)
- GrowMask (1)
- PreviewImage (12)
- MaskToImage (5)
- SolidMask (3)
- ImageToMask (3)
- ImageCompositeMasked (1)
- InvertMask (1)
- SamplerCustomAdvanced (2)
- BasicScheduler (2)
- ImageScaleToTotalPixels (1)
- RandomNoise (2)
- BasicGuider (2)
- VAEDecodeTiled (1)
- VAEEncode (1)
- UpscaleModelLoader (1)
- ImageUpscaleWithModel (1)
- KSamplerSelect (2)
- LoraLoaderModelOnly (2)
- LoadImage (3)
- DualCLIPLoader (1)
- CLIPTextEncode (1)
- ImageScale (1)
- SaveImage (3)
- StyleModelLoader (1)
- EmptyLatentImage (1)
- easy imageConcat (4)
- easy cleanGpuUsed (3)
- MaskPreview+ (2)
- GetImageSize+ (1)
- MaskToSEGS (1)
- DetailerForEach (1)
- LayerUtility: ColorImage (1)
- LayerUtility: ImageScaleByAspectRatio V2 (3)
- LayerUtility: ImageMaskScaleAs (1)
- LayerUtility: CropByMask (2)
- InpaintCrop (1)
- InpaintStitch (1)
- ShowText|pysssss (2)
- Load Image Batch (1)
Model Details
Checkpoints (0)
LoRAs (2)
flux\手部稳定器.safetensors
flux\洗图神器.safetensors