Skip to main content
Image Manipulation

Qwen Add Background

Remove white backgrounds from images and add realistic scenes using Qwen Image Edit Plus with LoRA

View details

Inputs

Loading input fields...
Execution Steps

Loading workflow structure...

Loading curated examples...

Overview

Qwen Add Background edits one or more images by removing a white or clean background and adding a prompted realistic scene behind the main object using Qwen Image Edit Plus with LoRA. Use it for product lifestyle concepts, ad mockups, ecommerce visuals, and campaign assets that need a richer setting.

Use cases

  • Place a product shot with a white background into a realistic campaign scene.
  • Create several product-background options for ecommerce or ad review.
  • Turn a clean cutout-style image into a richer lifestyle visual.
  • Use seed and image size controls when repeatable background exploration matters.

Input tips

  • Provide one or more public image_urls with white or clean backgrounds.
  • Use the prompt to describe the new scene, environment, lighting, surface, and mood.
  • Choose num_images from 1-4 when you want multiple background options.
  • Use image_size presets or custom dimensions only when the output canvas matters.
  • Set output_format to png, jpeg, or webp based on the final handoff need.
  • Adjust lora_scale, guidance_scale, inference steps, negative_prompt, safety checker, or seed only when needed.

Expected output

The AI Tool returns one or more edited images with downloadable URLs and optional content type, file name, file size, width, and height, plus the seed used and cost metadata. The shared image output view renders a single image or gallery and shows seed, image count, and the Qwen Add Background model label.

Caveats

  • Image URLs must be public and reachable.
  • The AI Tool works best on images with white or clean backgrounds.
  • Background replacement can change edges, shadows, reflections, product proportions, or fine details.
  • Generated product details, text, logos, and brand marks should be reviewed before use.
  • Safety filtering may block or alter requests that violate model rules.