FigureViewer

😩 Tired of jumping around on a paper when authors cite a figure two million pages away?
😊 Here is a Python utility that extracts images from PDF files and provides an interactive PyQt5-based GUI for viewing them.

Features

• Extract all images from PDF files automatically
• Navigate through images using mouse scroll or keyboard arrows

Requirements

• Python 3.7+
• pypdf
• PyQt5
• tqdm

Installation and usage

git clone [email protected]:tgrassi/figureviewer.git
pip install pypdf PyQt5 tqdm
python viwer.py path/to/your/file.pdf

Use mouse wheel to scroll or keyboard arrows.

Keys

• Mouse Wheel: Scroll up/down to move between images
• Arrow Keys: Use Up/Down/Left/Right to navigate
• PageUp/PageDown: Navigate images
• Space: Next image
• R: Refresh the current image
• T/A: Toggle always-on-top mode
• X: hide/show window content
• Home: Go to the first image
• End: Go to the last image
• Esc/Q: Quit the application
• H: Show help dialog

Use pre-extracted images (buffered mode):

python viewer.py path/to/your/file.pdf -b

Use mapping between image file and figure number:

Since not all the extracted files are relevant, or there are multiple images per figure, you can map the image file to the figure number.
For example, if Figure 1 corresponds to the second file (so you skip the first file), and Figure 3 is formed by three image files, you can create a file maps.txt

-1, 1, 2, 3, 3, 3, 4, 5, 6, 7

and run

python viewer.py path/to/your/file.pdf path/to/maps.txt

How It Works

1. Extracts images from each PDF page into an images/ folder
2. Saves images as image_XXXXXX.png with zero-padded numbering
3. Launches an interactive viewer displaying images one at a time
4. Window title shows current image number (e.g., "Fig. 1")