Run this workflow on InstaSD
Get started in minutes! Run this ComfyUI workflow online - no setup required.
Description
This workflow uses Grounding Dino to segment out the part of the original video to be replaced. We recommend a 48GB+ GPU for this workflow.
Models
| File | Destination | Source |
|---|---|---|
| AnimateLCM_sd15_t2v_lora.safetensors | /ComfyUI/models/loras/animatediff | Download |
| AnimateLCM_sd15_t2v.ckpt | /ComfyUI/models/animatediff_models | Download |
| control_v11f1p_sd15_depth.safetensors | /ComfyUI/models/controlnet | Download |
| epicphotogasm_ultimateFidelity.safetensors | /ComfyUI/models/checkpoints | Download |
| CLIP-ViT-H-14-laion2B-s32B-b79K.safetensors | /ComfyUI/models/clip_vision | Download |
| ip-adapter-plus_sdxl_vit-h.safetensors | /ComfyUI/models/ipadapter | Download |
| ip-adapter-plus_sd15.safetensors | /ComfyUI/models/ipadapter | Download |
| depth_anything_v2_vitl.pth | /ComfyUI/models/depthanything | Download |
Nodes
ADE_ApplyAnimateDiffModelSimpleADE_AnimateDiffSamplingSettingsADE_UseEvolvedSamplingADE_LoopedUniformContextOptionsRerouteLoraLoaderADE_LoadAnimateDiffModelCLIPTextEncodeControlNetLoaderAdvancedACN_AdvancedControlNetApplyIPAdapterCheckpointLoaderSimpleIPAdapterUnifiedLoaderDepthAnythingV2PreprocessorGroundingDinoModelLoader (segment anything)SAMModelLoader (segment anything)MaskToImageSetLatentNoiseMaskVAEEncodeGrowMaskWithBlurPrepImageForClipVisionPreviewImageLoadImageGroundingDinoSAMSegment (segment anything)KSamplerVHS_LoadVideoVAEDecodeVHS_VideoCombine