ComfyUI Extension: VRGameDevGirl Video Enhancement Nodes
Run ComfyUI workflows without the setup
No installs, no CUDA version roulette, no GPU sitting idle on your bill. Bring a workflow and run it in the browser.
Film grain and color match nodes designed for high-quality frame-by-frame video enhancement in ComfyUI.
Looking for a different extension?
Custom Nodes (164)
- BeatImpactAnalysisNode
- Beat-Aligned Scene Durations
- 🎨 Color Match To Reference
- 🎞️ Fast Film Grain
- 🌀 Fast Laplacian Sharpen
- 📏 Fast Sobel Sharpen
- 🎯 Fast Unsharp Sharpen
- Image From Folder (Index)
- Image From Folder (Index For Remake Mode)
- SRTLyricsMerger
- VRGDG_ArchiveLlmBatchFolders
- ✂️ VRGDG Audio Crop
- VRGDG_AudioCropTime
- VRGDG_AudioDelayByIndex
- VRGDG_BoxIT
- VRGDG Build Video Output Path (General)
- VRGDG Build Video Output Path (General_SRT)
- 🎧 VRGDG_CalculateSetsFromAudio
- VRGDG_CalculateSetsFromAudio_Queue
- VRGDG_CleanAudio
- 🌀 VRGDG_CombinevideosV2
- 🎬 VRGDG Combine Videos V3
- VRGDG_CombinevideosV5
- 🎞️ VRGDG_ConditionalLoadVideos
- VRGDG_CreateFinalVideo
- VRGDG_CreateFinalVideo_SRT
- VRGDG_CreateSilentAudio
- VRGDG Cycling Text Picker
- VRGDG_DisplayIndex
- VRGDG_DurationIndexFloat
- VRGDG Easy Multi Cycling Text Picker
- 🎞️ VRGDG_Extract_Frame_Number
- Face Fix - Composite Repaired Video
- Face Fix - Create Cropped Face Video
- Face Fix - Load Anchors (Meta Batch)
- Face Fix - Collect LTX Inputs
- Face Fix - Prepare Video and Anchors
- Face Fix - Store Enhanced Anchors
- 🧠 VRGDG General GGUF 🧠
- VRGDG_GeneralPromptBatcher
- 🧠 VRGDG General VLM 🧠
- VRGDG_GetAudioFilePath
- 📂 VRGDG Get File name Prefix
- 🔢 VRGDG Get Index Number
- VRGDG_GetRunIndexFromJson
- VRGDG_GetStems
- VRGDG_HumoReminderNode
- VRGDG_ImageIndex0HUMOEDIT
- VRGDG Image Switch (Index Map)
- VRGDG Image Switch (1-4)
- VRGDG Image Switch (Multi Dynamic)
- VRGDG Image Switch (Multi Dynamic 002)
- VRGDG_IndexedPromptChunker
- VRGDG_IndexedPromptChunkerV2
- VRGDG_IntToFloat
- VRGDG_IntToString
- VRGDG_Json2String
- VRGDG Latest SRT Auto Loader
- 🩺 VRGDG Llama CPP Doctor 🩺
- 🤖 VRGDG LLM Multi 🤖
- VRGDG_LLM_OutputSaver
- VRGDG_LLM_PromptBatcher
- VRGDG_LoadAudioFilePath
- VRGDG_LoadAudioSplitDynamic
- VRGDG Load Audio Split (General)
- VRGDG_LoadAudioSplit_HUMO
- VRGDG_LoadAudioSplit_HUMO_Transcribe
- 🗣️ VRGDG_LoadAudioSplit_HUMO_TranscribeV2
- 🎙️ VRGDG Load Audio Split HUMO Transcribe V3
- VRGDG_LoadAudioSplit_SRTOnly
- VRGDG_LoadAudioSplitUpload
- VRGDG_LoadAudioSplit_Wan22HumoFMML
- VRGDG_LoadAudioWithPath
- VRGDG_LoadLatestCombinedJsonText
- VRGDG_LoadText
- VRGDG_LoadTextAdvanced
- VRGDG_LoadVideos
- 💻 VRGDG Local LLM 💻
- VRGDG LTX Audio Only LoRA Train Chunk
- VRGDG LTX Audio Video LoRA Train Chunk
- VRGDG LTX IC Ingredients Grid
- VRGDG LTX LoRA Train Chunk
- VRGDG LTX Preview XYZ Plot
- VRGDG_LUTS
- VRGDG_LyricSegmentDurationMerger
- VRGDG_LyricSegmentJsonFixer
- VRGDG Lyric Segment Text Cleaner
- VRGDG_LyricsEmotionMerger
- VRGDG_MakeLUT
- 🌀 VRGDG_ManualLyricsExtractor
- Manual Lyrics Extractor (SRT Segments)
- VRGDG Multi Cycling Text Picker
- VRGDG_MultiStringConcat
- VRGDG_MusicVideoPromptCreator
- VRGDG_MusicVideoPromptCreatorV2
- 🌀 VRGDG_MusicVideoPromptCreatorJson
- VRGDG Musubi-Tuner Installer
- VRGDG_MuteUnmute4PromptCreatorWF_0
- VRGDG_MuteUnmute4PromptCreatorWF_1
- VRGDG_MuteUnmute4PromptCreatorWF_2
- 🚀 VRGDG NanoBanana Pro 🚀
- VRGDG_NoteBox
- VRGDG Optional Multi LoRA Model Only
- VRGDG Optional Multi LoRA Two Pass Strengths
- VRGDG_PadVideoWithLastFrame
- VRGDG Part 2 Workflow UI
- VRGDG Workflow 3 UI
- VRGDG_PostRunIndexStepper
- VRGDG_PromptCreatorUI
- VRGDG_PromptCreatorUI_V2
- VRGDG Prompt JSON Subject Prepender
- VRGDG_PromptMapJsonFixer
- VRGDG_PromptSpitterWithIndex
- ✂️ VRGDG_PromptSplitter
- VRGDG_PromptSplitter2
- VRGDG_PromptSplitter4
- VRGDG_PromptSplitterForFL
- VRGDG_PromptSplitterForFMML
- ✂️ VRGDG_PromptSplitterForManual
- VRGDG_PromptSplitter_General
- VRGDG_PromptSplitterJson
- VRGDG_PromptSplitterV2
- VRGDG_PromptTemplateBuilder
- VRGDG_PythonCodeRunner
- VRGDG_QueueTriggerFromAudio
- 🧠 VRGDG Qwen 2.5 🧠
- 🧠 VRGDG Qwen 3.5 🧠
- VRGDG_RunStateLogger_SRT
- VRGDG_SaveAudio
- VRGDG_SaveAudioFilePath
- VRGDG_SaveText
- VRGDG_SaveTextAdvanced
- VRGDG_SaveTextAdvancedConcat
- VRGDG_SetGroupStateMulti
- VRGDG_SetMuteStateMulti
- VRGDG_ShowAny
- VRGDG_ShowImage
- VRGDG_ShowText
- VRGDG_SmartSplitTextTwo
- VRGDG_SpeechEmotionExtractor
- VRGDG LTX Speed Character Lora Training
- VRGDG_SplitPrompt_T2I_I2V
- VRGDG_StoryBoardCreator
- VRGDG_StoryGroupJsonFixer
- VRGDG_String2Json
- VRGDG_StringConcat
- 🧠 VRGDG SuperGemma GGUF Chat 🧠
- Text to image prompts from concepts
- Text to video prompts from concepts
- VRGDG_TextBox
- VRGDG_ThemeSplitter
- ⏱️ VRGDG_TimecodeFromIndex
- VRGDG_TranscribeLyric
- 🎯 VRGDG Trigger Counter
- VRGDG_TrimFinalClip
- VRGDG_TrimImageBatch
- VRGDG_TrimImageBatch_SRTOnly
- VRGDG_UpdateLatestCombinedJsonPrompts
- VRGDG_UpdateLatestCombinedJsonPrompts_zimage
- VRGDG Video Folder Grid Plot
- VRGDG_VideoSplitter
- VRGDG VoxCPM2 Voice Clone / TTS
- VRGDG Z-Image LoRA Train Chunk
- VRGDG Z-Image Speed Character Lora Training
README
License
This ComfyUI custom node is licensed under AGPL-3.0.
You may not use this code in a closed-source paid app, private hosted service, SaaS product, or commercial wrapper unless you comply with the AGPL-3.0 license and provide the complete corresponding source code under the same license.
🎮 VRGameDevGirl’s Video / Image & HUMO Workflow Nodes for ComfyUI
Custom ComfyUI nodes for music video workflows and high-quality video/image enhancement.
Includes HUMO-based multi-scene workflows, audio splitting, video combining, and realtime-ready enhancement tools.
🎬 AI Music Video Workflow (ComfyUI)
Turn a reference image and an audio track into a fully AI-generated music video.
The workflow automatically splits lyrics, generates scene prompts, and syncs everything into a cinematic final video.
🚀 How it Works
- Upload a reference image (your main character).
- Import your audio file (with lyric syncing).
- Set a folder name for outputs.
- Customize the Prompt Creator to define style, mood, and scenes.
✨ Everything else — transcription, scene building, video chunks, and final rendering — runs automatically.
The result: a stylized, synced AI-driven music video.
🌟 Features
-
🎞️ Fast Film Grain (
FastFilmGrain)
Add controllable, grayscale or color grain for cinematic texture.
➕ Now includes abatch_sizesetting to reduce out-of-memory (OOM) issues on long or high-res videos. If you have a strong GPU, trybatch_size = 8. If you hit OOM, lower the value (e.g., to 4 or 2). -
🎨 Color Match to Reference (
ColorMatchToReference)
Align image tones to a reference image using LAB color matching.
➕ Also includes abatch_sizesetting to help manage VRAM usage. Start at 8 if your GPU allows — lower it if you get OOM errors. -
🎯 Fast Unsharp Sharpen (
FastUnsharpSharpen)
Simple and efficient sharpening using unsharp masking. -
🌀 Fast Laplacian Sharpen (
FastLaplacianSharpen)
Edge-based sharpening via Laplacian kernel for crisp detail. -
📏 Fast Sobel Sharpen (
FastSobelSharpen)
Gradient-based edge enhancement using Sobel filters. -
⚡ Optimized for image or video workflows and per-frame processing in ComfyUI.
-
💻 Designed to be efficient on modern GPUs.
📦 Installation
🧰 Using ComfyUI Manager (recommended)
- Open ComfyUI.
- Go to the Manager tab → Install Custom Nodes.
- Search:
vrgamedevor use this Git URL:https://github.com/vrgamegirl19/comfyui-vrgamedevgirl
🖐️ Manual Install
- Clone or download this repo to your
ComfyUI/custom_nodesdirectory. - Restart ComfyUI.
✨ Requirements
A requirements.txt file is now needed.
To install the required packages, run this from inside the python_embeded folder:
python.exe -m pip install -r ..\custom_nodes\comfyui-vrgamedevgirl\requirements.txt
requirements.txt (in the repo) includes:
kornia
librosa
imageio
🧠 Node Details
✅ Fast Film Grain (FastFilmGrain)
Adds customizable film grain to each frame.
Grain can be grayscale or saturated RGB, and sized to match your output resolution.
Inputs:
images: Frame tensor input.grain_intensity: Blend amount (0 to 1).saturation_mix: 0 = grayscale grain, 1 = full RGB noise.
🟢 Video-safe and very fast.
🎨 Color Match to Reference (ColorMatchToReference)
Matches image color distribution to a reference image using LAB space normalization.
Inputs:
images: Your video frames.reference_image: A single image to match tone and color against.match_strength: Blend between original and matched (0.0–1.0).
🎯 Fast Unsharp Sharpen (FastUnsharpSharpen)
Applies unsharp masking to enhance edges with a fast, low-cost blur pass.
Inputs:
images: Input image tensor.strength: Sharpening amount (0.0 to 2.0)
🚀 Lightweight and ideal for subtle sharpening.
🌀 Fast Laplacian Sharpen (FastLaplacianSharpen)
Enhances edges by applying a Laplacian kernel to bring out high-frequency detail.
Inputs:
images: Input image tensor.strength: Sharpening amount (0.0 to 2.0)
🧪 Gives a more "punchy" sharpen effect great for detail recovery.
📏 Fast Sobel Sharpen (FastSobelSharpen)
Uses Sobel filters to detect image gradients and amplify edge contrast.
Inputs:
images: Input image tensor.strength: Sharpening amount (0.0 to 2.0)
🧠 Sharpens by boosting directional edge response — great for outlines and detail clarity.
🛠️ Roadmap
- [x] 🎞️ Fast Film Grain (
FastFilmGrain) - [x] 🎨 Color Match To Reference (
ColorMatchToReference) - [x] 📏 Fast Sobel Sharpen (
FastSobelSharpen) - [x] 🌀 Fast Laplacian Sharpen (
FastLaplacianSharpen) - [x] 🎯 Fast Unsharp Sharpen (
FastUnsharpSharpen) - [ ] 🌫️ Local Contrast / Dehaze
- [ ] 🎛️ LUT Loader or Approximate Match
📁 Folder Structure
comfyui-vrgamedevgirl/
│
├── init.py
├── nodes.py
├── README.md
├── LICENSE
├── requirements.txt
└── .gitignore
🧑💻 Author
VRGameDevGirl
✨ Custom tools for cinematic AI workflows
💌 Questions or collabs? Reach out via GitHub
📜 License
This project is licensed under the MIT License.
Run ComfyUI workflows without the setup
No installs, no CUDA version roulette, no GPU sitting idle on your bill. Bring a workflow and run it in the browser.