1) First Time Video Tutorial :  https://www.youtube.com/watch?v=qczh3caLZ8o&ab_channel=JerryDavosAI

.

2) Animation with IP and Consistent Background Documented Tutorial :  https://www.patreon.com/posts/v3-0-bg-changer-96735652

.

3) Render Videos :  https://www.youtube.com/@Ai_Davos/shorts

.

HOW TO USE:

After you have rendered the raw Images from #2 Main Animation Raw file,

1) Enter the Paths in Purple Directory Nodes of the Raw Images from #2

2) Enter the Output path for saving the refined images.

3) Enter Batch Range

4) Overlapping Frames - 0 is Default (Put 5 or 10 for overlapping cross fade technique between batches)

5) Skip Frame is 0 by default (Increase it after every batch)

6) Batch Number Naming, increase it by 1 after every batch for organizing.

7) Refiner Upscale - Upscale value is 1.2 by default (Use 1.2 - 1.8 for best result for SD models)

8) See the Red note below for changing the refiner settings.

9) Enter Prompts for more accuracy.  

10) Choose Model checkpoint you want to use for refiner.

11) Loras work as usual (Some of not compatible with animatediff, test it first with 10 frames)

*Caution: Don't Use Lora on same or full strength, which was already used for making Raw Animation, it will double up.

12) After Rendering all the frames then Open the #4 AnimateDIff Face Fix if your faces are not looking good.

You can skip 12th step If faces are looking good or does not have faces in the video.

Tip - Overlapping workflow can be applied here in refiner pass, Example how to use:

Total Input Image: 350

Batch 1:  Range - 100 | Overlapping : 10 | Skip - 0

Batch 2:  Range - 100 | Overlapping : 10 | Skip - 100

Batch 3:  Range - 100 | Overlapping : 10 | Skip - 200

Batch 4:  Range - 50  | Overlapping : 0  | Skip - 300 ( Overlap is 0 for the last batch to avoid extra repetition)

Total Output Images: 380 (350 images + (10 x 3 images for each batch))

In Post fade the extra overlapping frames for smooth transition.

Tip 2 : You can use this refine workflow again and again till you get your desired details, with same upscaling value or higher, just don't use overlapping frames technique in refiner phases else it will mess up the controlnet images,

You can use overlapping technique in the last refiner phase.

Eg -

Raw phase > Refined Phase 1 > Refined Phase 2 > (Refined Phase 3 + Overlapping)

like this

Just update the Input Raw Images directory to Refined phase x directory and Output Node every time.  

Tip 3: This Workflow Can also be used as vid2vid style conversion, Just Input the Original Source Frames as Raw Input and Denoise upto 0.6 - 0.7.

And you can also the use these images for refiner again :D in Tip 2

____________________________________________________________________________________________________

3_0) AnimateDiff Refiner_v3.0

*Special Features are separated in different files due to technical limitation. All Files Download Folder Link : https://drive.google.com/drive/folders/1HoZxKUX7WAg7ObqP00R4oIv48sXCEryQ?usp=sharing

_________________________________________

If you need Cloud GPU : https://www.runcomfy.com/?ref=jerrydavos

My Discord Server : https://discord.gg/z9rgJyfPWJ