I recently stumbled upon a fascinating discovery in the world of image editing. QWEN Image Edit, a powerful tool, can extract depth and normal maps from images, and even use them like controlnets – without requiring Lora. This is a game-changer for creatives and developers alike.
I was experimenting with turning people into LEGO figures when I accidentally stumbled upon a depth map. To my surprise, I could recreate it with any image using a simple prompt. The tool can also generate normal maps, which seem to produce similar results to dedicated normal map estimators like Metric3D.
But that’s not all. This feature is reversible, meaning you can take any depth map and use it to generate an image. It’s like having a built-in controlnet.
I’ve tested the workflow myself, and I’m excited to share the results. You can try it out for yourself using the workflow I’ve shared on Pastebin.
The Implications Are Huge
This feature opens up new possibilities for creatives and developers. Imagine being able to generate detailed 3D models from 2D images, or creating realistic environments with ease. The possibilities are endless.
How It Works
The process is surprisingly simple. QWEN Image Edit uses advanced algorithms to extract depth and normal maps from images. These maps can then be used to generate new images or even 3D models.
What This Means for the Future of Image Editing
This feature is a significant step forward for image editing. It democratizes access to advanced 3D modeling tools, making it possible for creatives to produce high-quality work without requiring extensive training or expensive software.
Try It Out for Yourself
If you’re as excited about this feature as I am, give it a try. You can find the workflow on Pastebin, and start experimenting with depth and normal maps today.