Launch Image Object Remover →
Opens in current tab
Image Object Remover
A client-side tool for removing unwanted elements from images. Paint over logos, text, watermarks, or other objects and the tool fills in the area using surrounding pixels.
Features
Image Input
- Drag & Drop: Drop images directly onto the page
- File Picker: Click to select images from your device
Mask Painting
- Adjustable Brush: Slider to control brush size (5–80px)
- Visual Preview: Brush cursor preview follows your mouse
- Red Overlay: Painted mask is shown as a semi-transparent red highlight
- Undo: Step back through brush strokes
- Clear Mask: Reset the mask and start over
Inpainting Algorithm
- Onion-Peel Fill: Fills the masked region from the outside in, layer by layer, using only known (non-masked) pixel values as sources
- Smoothing Pass: Applies additional averaging passes to eliminate banding artifacts
- Progress Indicator: Visual progress bar during processing
Export
- PNG Download: Save the result as a PNG file
Usage
- Open the tool and upload an image
- Adjust the brush size as needed
- Paint over the element you want to remove
- Click “Remove Selection”
- Download the result
Limitations
This uses a simple client-side diffusion algorithm. It works best for:
- Small to medium elements on relatively uniform backgrounds
- Logos, watermarks, and text on gradients or solid colors
For complex scenes or large removals, an AI-based inpainting service would produce better results.
Technical Implementation
- No Dependencies: Single HTML file with embedded CSS and JavaScript
- Canvas API: Separate canvases for the image and mask overlay
- Client-Side Processing: All computation runs in the browser — no server or API calls
- Non-blocking: Uses
requestAnimationFrameto keep the UI responsive during processing