PHOTOGRAPHY

Make your face appear in any photo.

Upload face pict, generate and it will be full bodybase model https://tensor.art/models/717542486853079143/CND_AK_Model-1Include IPAdapter and HiresFixAnother Model https://tensor.art/u/676685502350362684/modelsEnjoy

Note:To generate the best results, make the height and width atleast 896x896 size and steps minimum of 45 (the more details in prompt the higher the steps should be).Instruction for best results:On PC:Checkpoint - must be SDXL only or it won't work. Choose your best preference.Load Input/Base Face Image - best image quality equals best results.Pose Image Section - just upload a FULL BODY photo of a person if you want results that are not portrait.Steps - more steps gives more details, but it will take more credits. Suggested minimum steps is 15, it's up for you to decide. If your prompt has a lot of details better increase steps, I personally put it between 45 and 60.Height & Width - For best results, put it to 1080x1080.On mobile:First (1) upload image : Pose Image Section - just upload a FULL BODY photo of a person if you want results that are not portrait.Second (2) upload image : Load Input/Base Face Image - best image quality equals best results.I know this is more costly than the first face swap but I'm looking for the easiest way but cheaper workflow that gives great results.

NoteI added POSE IMAGE for those who want to generate photos that is not portrait and is half-body shot.Upload any photo on to POSE IMAGE that captured in half-body and you'll get half-body photo results.Thank you for the support.Generate Ai Art with Your FACE or Anyone's Face you like!From Medieval Time to Vintage to Futuristic Ai photo.Show them to your friends, and surprise them.

Highly detailed image is generated for a beautiful young girl. Wow~ Radio Button prompt input is employed now!!

Schnell variant of Flux.1 by Black Forest Labs.Lighter, faster, cheaper.At a compromise of course.Usual recommended step number is 4, but since it takes the same amount of credits I've set it to 5. A higher number may still enhance the quality.Recommended samplers and schedulersFor starters, euler is a safe choice most of the time. But you might want to experiment a bit on the topic.Flux.1 likes deterministic samplers and is not able to converge stochastic ones. Which means, AVOID sde or ancestral samplers as they will only give you totally blurred out images.Also avoid ddpm which bevahes similarly to euler_ancestral.The remaining ones, most of them should work fine.Simple and sgm_uniform schedulers work fine with probably all the samplers.Karras and exponential can't converge well and give "foggy" images which you can still use in a creative way sometimes.Normal is fine with euler and others, but can get oversharpened with dpmpp_2m or heun.Beta sometimes can get harsh but seemed to be more versatile than normal.

Create real and beautiful image with your face！

!Lưu ý: Nếu thừa ánh sáng hãy giảm gradient Light Maskhoặc có thể bỏ bớt prompt có miêu tả về ánh sáng.Workflow hoạt động tốt với ảnh chân dung và ảnh full body, có thể không tốt nếu dùng ảnh quá nhiều chủ thể trong hình ảnh,Nếu bạn thấy tools AUTO REMOVE BACKGROUND HƠI TỆ CÓ THỂ DÙNG TÁCH NỀN BẰNG PTSHoạt động tốt với ảnh kích thước lớn, nếu ảnh kích thước quá 2K pixel, sẽ tự động giảm xuống 2K và bước cuối chạy upscale lên 4K.CHÚC CÁC BẠN TRẢI NGHIỆM VUI VẺ!

Simple and low cost img2img for daily jobs. Now upgraded to SD3. The quality got better.

Before using, please note the following:- You need to paint a small mask area on the object you want to transform- Enter a prompt (segment anything) to identify the object you have painted the mask- Enter a prompt to describe what you need to changeI wish you a good experience with my AI toolWorkflow build by SoraSleep (SoraSleep.com) blogger, AI model creator