中 Download
Features Formats Shortcuts Tech Docs GitHub Download

Region screenshot, OCR and color picker

Capture part of the virtual desktop, recognize text locally with PaddleOCR, and pick pixel colors — all offline.

Updated 2 min read Edit this page on GitHub
All documentation
All documentation
Getting started
Getting started Spacebar preview Keyboard shortcuts
Core features
Editing and saving Offline voice input Region screenshot, OCR and color picker
Reference
Settings and system integration Technology and project layout Privacy and data Troubleshooting Frequently asked questions
On this page

SigXor's screenshot flow keeps selection, preview, copy, save, OCR and color picking in one place. Everything runs locally.

Start a screenshot#

Press Alt + </kbd> by default. In **Settings → Voice & screenshot** you can change the modifier to Ctrl, Ctrl + Shift or Win, or enable **No modifier ** for laptop keyboards that need Fn to produce the backtick.

When the screenshot starts, SigXor freezes the entire virtual desktop and hides its own windows:

Action Behavior
Drag Select a region
Double-click Capture the entire virtual desktop
Enter Confirm the selection
Esc / right-click Cancel
Minimum selection 4 × 4 pixels

Preview and toolbar#

The captured region is shown at 1:1 physical pixels and can be dragged. The toolbar follows the preview and offers:

  • OCR: recognize text in the capture.
  • Color picker: enter the pixel color picker.
  • Copy image: put the image on the Windows clipboard for pasting into chat or documents.
  • Save: write a PNG to Documents/FNSoftware/SigXor/Screenshots by default.
  • Done: closes the screenshot flow only and does not copy. Use Copy image to copy.

OCR#

OCR uses RapidOcrNet with PaddleOCR locally:

  • On first use, PP-OCRv6 downloads on demand (about 30 MB); if that fails, SigXor falls back to PP-OCRv5 (about 16 MB).
  • A result window opens immediately with progress, then shows read-only text.
  • Character and line counts are displayed; copy the text or close the window.
  • Closing the result window ends the screenshot flow.

Accuracy depends on sharpness, font, size, language and background contrast. Capture at a larger size rather than scaling up a blurry image.

Color picker#

Choose Color picker from the screenshot toolbar for a Snipaste-style picker:

  • Frozen screen plus a 15 × 15 pixel magnifier showing the current HEX / RGB value.
  • Move the mouse to sample; click or press C to copy.
  • Press Shift to switch HEX / RGB.
  • Use WASD or the arrow keys to nudge one pixel at a time.
  • Esc or right-click exits and restores the screenshot preview with a confirmation that the color was copied.

Multiple monitors and DPI#

  • The screenshot covers the whole virtual desktop, including negative-coordinate layouts.
  • The overlay is currently a single virtual-desktop window. Multiple monitors with the same DPI work normally. Mixed DPI layouts are still being validated on real hardware; setting all monitors to the same scale is a reliable workaround.
  • SigXor's own windows are hidden during capture and restored afterwards.

Privacy#

Screenshots, OCR and color picking all happen on-device. A screenshot is written to disk only when you choose Save; nothing is uploaded automatically.

Troubleshooting#

  • Hotkey does nothing: make sure Enable region screenshot is checked and the modifier does not conflict with the voice hotkey.
  • Fn keyboards: enable No modifier `.
  • OCR returns nothing on first use: check the network, wait for the model download, or download it manually from Services & models.

See Troubleshooting for more.