Tem um workflow do ComfyUI? Gere uma documentação legível direto no navegador.Torne um workflow legível.Experimentar agora ↗
Workflow for creating a morphing effect between two images (or video frames). It uses a combination of RiFE frame interpolation and i2i with IPA (SD1.5).
Perfil de execução
Descrição da fonte
Workflow for creating a morphing effect between two images (or video frames). It uses a combination of RiFE frame interpolation and i2i with IPA (SD1.5).
The main difference between this and pure RiFE is that using diffusion model (SF1.5) for key frames allows a much smooth effect where in-between images make more sense, rather than being gargled blobs of color.
Using SD1.5 adds a bit more otherworldly mesmerizing effects due to being trained on low resolution (512), and key frames being scaled to large resolution (~1024).
For examples of this approach being used for cuts between clips, I have used this on a youtube series here:
Comentário gerado por IA
Explicação gerada por IA com base nos detalhes da fonte e da configuração. As sugestões são claramente identificadas.
Frame morphing creates a morphing effect from two images or video frames by combining RiFE frame interpolation with image-to-image processing and IPA on SD 1.5, producing in-between images.
It combines diffusion-based key-frame processing with RiFE frame interpolation for the in-between images.
The v3.0 listing includes alternatives for morphing back and forth between two or three images.
Use two images or two video frames as inputs.
The output is a morphing effect containing in-between images.
Open the v3.0 Civitai listing at https://civitai.com/models/1656349?modelVersionId=2004093; it is published by Hypersphere.
Listed model files are BearlySD.safetensors, clip_vision_h.safetensors, dreamcolor.safetensors, ip-adapter_sd15.safetensors, juggernaut_reborn.safetensors, and thickline.safetensors.
Listed node packs include ComfyLiterals, comfyui_essentials, comfyui-frame-interpolation, comfyui-image-selector, and comfyui-kjnodes.
Other listed node packs are comfyui-various, comfyui-videohelpersuite, masquerade-nodes-comfyui, mikey_nodes, and comfyui_ipadapter_plus.
The listed base model is SD 1.5.
The description mentions SD 1.5 training at 512 and key frames scaled to about 1024.
Before use, verify that BearlySD.safetensors, clip_vision_h.safetensors, dreamcolor.safetensors, ip-adapter_sd15.safetensors, and thickline.safetensors are available under those exact names.
Sugestão · não verificado
Before use, confirm which installed node pack provides comfyui_ipadapter_plus.
Sugestão · não verificado
Isso precisa ser editado?
Faça login para enviar uma solicitação de edição.
Fontes
1 fonteTrechos de fonte
2 trechosSource context: 547 downloads · Type Workflows · Base model SD 1.5
Estimativa de requisito VRAM
Estimativa indisponível
2,03 GB distribuídos em 2 de 7 arquivos de modelo. Total de arquivos do modelo + 25% de overhead de carregamento + 2 GB de buffer de execução, arredondado para cima.
Requisitos
Requisitos 17BearlySD.safetensors
Não resolvidoLoRA · Unknown
Vision model · Civitai · 1329096
Fontes possíveis
dreamcolor.safetensors
Não resolvidoLoRA · Unknown
Checkpoint · 42.6 MB · Hugging Face · h94/IP-Adapter
Checkpoint · 1.99 GB · SAFETENSOR · fp16 · Civitai · 46422
Checkpoint · Hugging Face · VMTamashii/rife49
thickline.safetensors
Não resolvidoLoRA · Unknown
Pacote de nós · Registry
Pacote de nós · Registry
Source context: 513 downloads · Type Workflows · Base model SD 1.5
Pacote de nós · Registry
Pacote de nós · Registry
Pacote de nós · Registry
Pacote de nós · Registry
Pacote de nós · Registry
Pacote de nós · Registry
Pacote de nós · Registry
Pacote de nós · Registry
