🚀 PERFECT Background Remover | InspyreNet | One-Click | Zero Setup | ComfyUI

详情

模型描述

🚀 PERFECT Background Remover | InspyreNet | One-Click | Zero Setup | ComfyUI


⚡ What This Does

Instantly removes backgrounds from ANY image with surgical precision — no manual masking, no Photoshop, no subscription fees. Just drop your image, hit queue, and get a clean transparent PNG in seconds.

This workflow uses InspyreNet — one of the most accurate open-source background removal models available, outperforming many paid tools like Remove.bg on complex edges (hair, fur, transparent objects).


🎯 Why You NEED This Workflow

Feature This Workflow Remove.bg Photoshop Cost FREE $9+/month $55+/month Privacy 100% Local Cloud (your images uploaded) Local Batch processing Unlimited Limited by plan Manual Edge quality Excellent Good Manual effort Setup time ~2 minutes Instant (paid) Already installed


📦 Required Custom Nodes

Install via ComfyUI Manager → Search & Install:

comfyui-inspyrenet-rembg

One node only. No model downloads required — InspyreNet downloads automatically on first run (~170MB).


🔧 Workflow Overview

[Load Image] ──► [InspyreNet Rembg] ──► [Save Image]
     │                   │                     │
  Any image       AI Background            Transparent
  JPG/PNG/WEBP    Removal Engine           PNG output

3 nodes. Dead simple. Incredibly powerful.


🖼️ Perfect For

  • Product Photography — Remove cluttered backgrounds for e-commerce

  • Portrait Cutouts — Clean subject extraction with hair detail preserved

  • AI Image Compositing — Cut subjects from generated images for new scenes

  • Graphic Design Assets — Batch-process hundreds of images overnight

  • Sticker Creation — Perfect transparent cutouts for merch, prints, stickers

  • Game Assets — Isolate characters/objects from reference photos

  • Video Thumbnails — Quick subject isolation for YouTube/social media


📥 Installation & Setup

Step 1 — Install ComfyUI Manager

If you don't have it: ComfyUI Manager GitHub

Step 2 — Install the Custom Node

  1. Open ComfyUI Manager → Install Custom Nodes

  2. Search: inspyrenet

  3. Install: comfyui-inspyrenet-rembg

  4. Restart ComfyUI

Step 3 — Load the Workflow

  1. Download the .json file below

  2. Drag & drop it into ComfyUI OR use Load button

Step 4 — Run It

  1. Click the Upload Image button on the Load Image node

  2. Select your image

  3. Hit Queue Prompt

  4. Find your result in ComfyUI/output/ folder

First run: InspyreNet will auto-download (~170MB). Wait ~30-60 seconds. Subsequent runs: Near-instant processing.


⚙️ Settings & Optimization

torchscript_jit Setting (on InspyreNet node)

Setting Speed Quality Use When default ⚡⚡⚡ Fast ★★★★★ Best Recommended for most users on ⚡⚡⚡⚡ Fastest ★★★★ Great When processing large batches

Pro Tips for Best Results

🔹 Input quality matters — Higher resolution input = sharper edges on output
🔹 Contrast helps — Images with clear subject/background contrast get the cleanest cuts
🔹 Works great on: people, products, animals, objects, logos
🔹 Challenging cases: glass, water, smoke, very similar colors to background
🔹 For challenging cases: Use the MASK output to refine in your preferred node (e.g., connect to a mask blur or manual paint node)


🔄 Extending This Workflow

Want to do more? Here are popular extensions:

➕ Add a Custom Background Color Connect output → Image Composite node with a solid color

➕ Add a New Background Image Connect output → Image Composite → Load second background image

➕ Batch Processing Replace Load Image with a batch loader to process entire folders

➕ Save with Custom Filename Change the filename_prefix field on the Save node to organize your outputs


🖥️ System Requirements

Minimum Recommended GPU VRAM 2GB 4GB+ RAM 8GB 16GB+ ComfyUI Version 0.9.0+ Latest Python 3.10+ 3.11+ OS Win/Mac/Linux Any

CPU Mode: Works but significantly slower (~10-30s per image vs <1s on GPU)


❓ Troubleshooting

"InspyrenetRembg node not found" → Install the custom node via ComfyUI Manager and restart

"Model download failed" → Check internet connection. The model downloads from HuggingFace on first run

"Output has rough edges" → Try higher resolution input. Add a Gaussian blur node to the MASK output for softer edges

"Black background instead of transparent" → Make sure you're saving as PNG (not JPG). JPG doesn't support transparency


📊 Performance Benchmarks

(Tested on RTX 3080 10GB)

Image Resolution Processing Time 512×512 ~0.3s 1024×1024 ~0.8s 2048×2048 ~2.1s 4096×4096 ~6.4s


🤝 Credits & Acknowledgements

  • InspyreNet Model — State-of-the-art salient object detection network

  • comfyui-inspyrenet-rembg — ComfyUI integration node

  • ComfyUI — The best open-source AI image generation framework


💬 If this workflow helped you, sharing your results in the comments

means the world! And if you'd like to support future workflows,

the Buzz button is right there 😊


Workflow Version: 1.0 | ComfyUI: 0.9.2+ | Node: comfyui-inspyrenet-rembg v1.1.1


🏷️ Tags

background-removal rembg inspyrenet comfyui-workflow transparent-background product-photography portrait cutout compositing free-tool no-subscription batch-processing utility0.

此模型生成的图像