Qwen-Image Edit - Virtual Staging Any Furniture using ComfyUI - Full Tutorial

2D to 3D Furniture Placement in ComfyUI | Realistic AI Interior Design Tutorial Transform simple 2D images of furniture into high-quality, 3D-integrated interior renders! In this video, I walk you through my custom ComfyUI workflow for taking a flat image of a couch or TV stand and placing it realistically into any room with consistent lighting, shadows, and textures. I spent over a week perfecting this pipeline, combining the latest in vision-language models and 3D generation to get professional-level results. What you’ll learn in this tutorial: ✅ How to convert 2D furniture images into 3D models (.glb) using Trellis.2 by Microsoft. ✅ Setting up the Qwen image Edit 2511 and Union-DiffSynth nodes and lighting loras in ComfyUI. ✅ Utilizing Normal Maps and Masks for precise object placement and light interaction. ✅ Applying high-quality textures from reference images to your generated 3D models. ✅ A step-by-step breakdown of the ComfyUI nodes, including the use of Lightning LoRA for faster inference. ✅ How to sequentially add multiple items to a scene using upscaling techniques (SEED V2). Tools & Models Used: ComfyUI: The ultimate node-based interface for stable diffusion. Trellis.2 (Microsoft): For high-fidelity 2D-to-3D generation. Qwen-VL-Chat-GGUF: Advanced vision-language model for image editing. Union-DiffSynth: For effective control over normal maps and depth. Lightning LoRA: For high-quality results in fewer steps. SEED V2 Node: For high-quality image upscaling between renders. PC Specs: Gpu: Nvidia RTX 5060 Ti 16 GB : https://amzn.to/4rU7xRy Ram: 64gb 4x16gb Kingston Fury : https://amzn.to/473HoaG Model Used : Q4_K_M v 2511 gguf Pro Tip: High-quality source images are key! The better your initial photo, the more realistic your final 3D render will look. If you found this workflow helpful, don’t forget to Like, Subscribe, and Hit the Notification Bell for more deep dives into AI-powered design! ig :   / kintugk   x : https://x.com/gk_kintu Timestamps: 0:00 - Intro & Result Showcase 0:50 - Preparing Your 3D Assets (Trellis.2) 2:29 - ComfyUI Workflow Overview 5:41 - The Power of Normal Maps 10:04 - Understanding the VAE Encoder 12:24 - Prompting for Realistic Shadows & Textures 14:21 - Adding Multiple Objects & Upscaling 16:23 - Final Thoughts & Outro #ComfyUI #AIGeneratedArt #InteriorDesign #3DModeling #QwenVL #StableDiffusion #Trellis3D #AIWorkflow #DigitalArtTutorial #MicrosoftTrellis