Upscale images and remove backgrounds — locally on Windows.
Process one image or an entire batch. Increase real pixel dimensions up to 4×, create transparent PNGs, and keep image processing on your PC.
Real Resolution Multipliers
Increase actual pixel dimensions up to 4× (16× total pixel count) with dedicated photo and illustration AI models.
Transparent PNG Cutouts
Isolate subjects from backgrounds with optional edge refinement for hair and fur, outputting clean transparent PNGs.
Independent Batch Queues
Drop multiple files, configure destination folders, process batches in one go, and selectively rerun single items.
On-Device Processing
AI computation happens on your PC hardware. Your image files are not uploaded for processing.
More pixels, not just a sharpening filter.
At 4×, an 800 × 800 image is exported at 3200 × 3200 — four times the width and height, or 16× the total pixel count. PixelUp uses AI upscaling while increasing the actual image resolution.
- Real x2 and x4 output dimensions
- Photo and illustration models
- Batch queue
- Full-size before/after comparison
- Individual image reruns
- Multiple output formats
Full-Size Before / After Inspector
Examine texture, edge sharpness, and color fidelity with an interactive split divider, pan, and zoom.
Whisker & Fur Edge Details
Inspect clean cutouts against the authentic transparency checkerboard.
Transparent cutouts without uploading your images.
Remove backgrounds from one image or an entire batch, refine difficult edges such as hair or fur, compare the result, and export transparent PNG files.
- Transparent PNG output
- Batch processing
- Optional edge refinement
- Before/after preview
- Individual image reruns
- Local image processing
Built for repetitive image work.
Each PixelUp tool has its own focused workspace and queue. Add multiple files, choose an output folder, process the batch, and rerun individual images when needed.
Independent Workspaces
Upscale and Remove Background operate as completely independent tools. You choose whichever tool you need for the job without being locked into an automated or unwanted pipeline.
Selective Reruns & Removal
Queue dozens of images at once. Pause the queue at any point, clear finished items, or trigger a rerun on a single specific image with adjusted parameters.
Safe output handling
Choose any target folder on your PC. PixelUp writes temporary files during inference, ensuring only fully processed, complete images appear in your destination folder.
Your images stay on your computer.
Image processing runs locally on your Windows PC. The background-removal model may be downloaded automatically on first use, but your image files are not uploaded for processing.
PixelUp uses compatible GPU acceleration when available via DirectX 12 and DirectML. If your hardware encounters a driver reset or does not support DirectX 12, PixelUp automatically steps down tile sizes or falls back safely to CPU execution.
Zero Image Uploads
PixelUp does not upload your image files or pixel data for processing.
No per-image cloud credits
Local processing does not consume per-image cloud processing credits. PixelUp is planned as a $19.99 USD one-time purchase with no subscription.
One-time model download
Upscale models are included with the application. The BEN2 background-removal model (~223 MB) downloads from Hugging Face on first use, after which it runs locally.
How it works
Designed with native Windows ergonomics for speed, clarity, and control.
Choose a tool
Select either Upscale or Remove Background from the streamlined sidebar navigation depending on your current task.
Add your images
Drag and drop single files or entire batches directly into the drop zone, or use the Windows native file picker. Supports PNG, JPG, WebP, BMP, and TIFF.
Process and compare
Start the queue with one click. Inspect completed results side-by-side using the built-in split divider, pan, and zoom.
System requirements
PixelUp is optimized specifically for modern 64-bit Windows systems.
| Operating System | Windows 10 version 2004 / Build 19041 or later, or Windows 11. The initial public release is planned for x64 systems. |
|---|---|
| Graphics | A DirectX 12-compatible NVIDIA, AMD, or Intel GPU is recommended for faster processing through DirectML. PixelUp can use a slower CPU fallback after supported GPU execution failures. |
| Processor | 64-bit Intel or AMD processor. |
| Memory | Memory requirements depend on the source image dimensions and selected scale. PixelUp checks available memory before processing very large images. |
| Storage | Storage is required for the application, exported images, and an additional background-removal model of approximately 223 MB. Final installed size will be published with the release package. |
| Internet | Upscale models are included with the application. Remove Background requires an internet connection the first time its model is downloaded; subsequent image processing can run locally. |
Frequently asked questions
Accurate and factual answers about how PixelUp operates on your system.
Does PixelUp upload my images?
No. All image upscaling and background-removal inference runs locally on your PC hardware. Your images and pixel data are never uploaded to any cloud server or remote service for processing.
Does upscaling really increase the image dimensions?
Yes. PixelUp increases the actual pixel count and canvas dimensions. For example, selecting 4× on an 800 × 600 input produces a true 3200 × 2400 image file (4× width and 4× height, multiplying the total pixel count by 16×). It is not a cosmetic sharpening filter or screen-magnification trick.
What is the difference between x2 and x4?
x2 doubles both width and height (yielding 4× total pixels) using the Photo x2 model, which provides a gentler upscale ideal when 4× would produce an excessively large file. x4 quadruples width and height (yielding 16× total pixels) and is available in both photographic and illustration/anime models.
Can I process several images at once?
Yes. Both Upscale and Remove Background feature a dedicated batch queue. You can drag and drop multiple files at once, select all or individual items, process the entire queue, and rerun specific images if you want to test different settings.
Which image formats are supported?
For input, PixelUp supports PNG, JPG, JPEG, WebP, BMP, and TIFF. For output, the Upscale tool supports exporting to PNG, JPEG, and WebP. The Remove Background tool exports transparent cutouts as PNG files.
Is a powerful GPU required?
A dedicated DirectX 12 compatible GPU (NVIDIA, AMD, or Intel) with DirectML support provides the fastest processing speeds. However, a high-end card is not strictly mandatory: PixelUp features automatic dynamic tile sizing and includes a resilient CPU fallback mode to execute inference if a compatible GPU is unavailable or busy.
Does background removal work with hair and fur?
PixelUp is designed to preserve fine edges such as hair, fur, and whiskers. Results vary with lighting, contrast, and image complexity. The optional Refine edges setting can improve the appearance of difficult boundaries.
How will the free trial work?
At launch, PixelUp will include 10 successful Upscale image exports and 10 successful Remove Background image exports. The limits are independent. Failed or cancelled jobs will not count, while every successfully exported image—including a rerun—will count once.
Is PixelUp a subscription?
No. PixelUp is planned as a $19.99 USD one-time purchase with no recurring subscription.
Where will PixelUp be available?
PixelUp is planned for distribution through the Microsoft Store. The public Store listing is not available yet.
Does PixelUp work offline?
Image and pixel processing runs locally on your PC. Upscale models are bundled directly with the application and operate locally from installation. The Remove Background tool requires an internet connection on its first run to download the open-source BEN2 model (~223 MB); after that, its image processing runs locally. A commercial release may require an internet connection for purchasing, license activation, or software updates. Your images and pixel data are never uploaded or sent to any payment provider.
Which versions of Windows are supported?
PixelUp supports 64-bit Windows 10 (version 2004 / Build 19041 or higher) and Windows 11 (x64). 32-bit (x86) and legacy Windows releases (Windows 7/8) are not supported.