SlideCraft

A full-featured web-based slide editor with OCR text editing, AI-powered logo removal, shape tools, image filters, video processing, templates, version history, and more.

What is SlideCraft?

SlideCraft is a web-based PowerPoint slide editor built for editing image-based presentations (like those generated by NotebookLM, Gamma, or similar AI tools). It lets you:

Edit text directly on slide images using EasyOCR detection
Remove NotebookLM watermarks from slides and videos seamlessly
Draw shapes, freehand, add images with a full creative toolkit
Apply image filters — brightness, contrast, blur, grayscale, sepia
Manage versions and templates — save, restore, reuse
Process videos — remove logos with real-time progress and stop control
Export as PPTX, PDF, PNG ZIP, or animated GIF

Built with Flask + vanilla JS. No React, no build step. Just python app.py and go.

Screenshots

Upload your PPTX, edit text with OCR, remove logos, export clean slides.

Features at a Glance (40+)

Drawing & Shape Tools

Feature	Description
Text Overlays	Styled text boxes with full typography controls
Rectangle	Colored rectangles with fill and opacity
Circle / Ellipse	Draw circles and ellipses
Line / Arrow	Draw lines with arrowheads
Callout	Speech bubble callout shapes
Freehand Drawing	Pen tool — draw anything with mouse
Cover & Type	Draw over text to cover + replace it
Image Overlays	Upload and place images on slides

Text Editing & OCR

Feature	Description
EasyOCR Detection	One-click text detection — no external binaries needed
Click to Edit	Click any detected text region to replace it
Auto Color Match	Samples background + text color from the original image
Auto Font Weight	Detects normal/semibold/bold/extrabold via ink density analysis
Auto-Fit Text	Text auto-shrinks to fit within the bounding box
Inline Editing	Double-click any text to edit directly on the canvas
Apply / Bake	Burns all edits permanently into the slide image

Typography Controls

Feature	Description
Font Family	12 system fonts: Arial, Segoe UI, Calibri, Verdana, Georgia, Tahoma, etc.
Font Size	8-72pt with precise numeric control
Bold / Italic / Underline	Full text style controls
Text Alignment	Left, center, right
Letter Spacing	-5 to 20px tracking adjustment
Line Height	0.8x to 3x multiplier
Text Transform	None, UPPERCASE, lowercase, Title Case
Text Shadow	Drop shadow with custom color picker
Text Outline	Stroke outline with custom color
Bullet / Numbered Lists	Toggle list formatting

Image Filters

Filter	Description
Brightness +/-	Adjust slide brightness
Contrast +/-	Adjust slide contrast
Saturation +/-	Adjust color saturation
Blur	Gaussian blur effect
Sharpen	Sharpen details
Grayscale	Convert to black & white
Sepia	Vintage warm tone
Reset	Revert to original

Logo & Watermark Removal (Slides)

Feature	Description
One-Click Logo Removal	"Remove Logo" strips NotebookLM watermark from all slides
Auto Watermark Detection	Scans slide corners for watermarks/logos using edge analysis
Remove from Single Slide	Remove detected watermark from current slide only
Remove from All Slides	Remove detected watermark across all slides at once
Confirmation Dialog	Delete confirmation before any destructive watermark removal
OpenCV Inpainting	Telea algorithm — samples from all surrounding edges seamlessly

Custom Watermark (Add Your Own)

Feature	Description
Text Watermark	Add text like "CONFIDENTIAL", "DRAFT", custom text
Image Watermark	Upload a logo/image as watermark overlay
Position Control	Center, Top-Left, Top-Right, Bottom-Left, Bottom-Right, Tiled
Opacity Control	5% to 80% transparency
Scale Control	Image watermark size as % of slide width
Apply to All	Watermark applied to every slide at once

Video Logo Remover

Feature	Description
Any Format	MP4, MOV, AVI, MKV and more
Any Length	Handles 3-8+ minute videos
Visual Region Select	Drag a resizable box over the logo on a preview frame
OpenCV Inpainting	Same seamless per-frame removal as slides
Audio Preserved	Audio track muxed back via MoviePy
Real-Time Progress	Live frame count, percentage, ETA, and elapsed time
Stop Button	Cancel processing mid-way — cleans up temp files
Auto Logo Detection	Automatic corner analysis to suggest logo position
Replace Logo	Swap the logo with your own image instead of removing
Video Trimming	Cut start/end before processing
Batch Video Upload	Upload multiple videos at once

Editing Tools

Feature	Description
Resize Handles	8-point resize on any selected overlay
Drag & Reposition	Move overlays by dragging
Undo / Redo	`Ctrl+Z` / `Ctrl+Y` with 50-step history per slide
Copy / Paste	`Ctrl+C` / `Ctrl+V` across slides
Snap Grid	Toggle alignment grid (every 10%) with 5% snapping
Slide Reordering	Drag thumbnails to rearrange
Crop Tool	Crop slide images to a selected region
Rotate	Rotate slides 90 CW / CCW
Right-Click Menu	Context menu: Edit, Duplicate, Delete, Bring to Front, Send to Back
Keyboard Shortcuts	Press `?` to see the full shortcut cheatsheet

Insert Tools

Feature	Description
QR Code Generator	Enter a URL, generates QR code as an image overlay
Custom Watermark	Add text watermark — center, bottom-right, or tiled, with opacity control
Find & Replace	Search and replace text across all slide overlays

Templates & Version History

Feature	Description
Save Template	Save current slides + overlays as a reusable named template
Load Template	Restore slides from a saved template
Delete Template	Remove templates you no longer need
Save Version	Snapshot current state for rollback
Restore Version	Revert to any previously saved version
Version List	Browse all saved versions with slide count

Comments & Annotations

Feature	Description
Comment Tool	Click anywhere on a slide to place a comment pin
Comment Panel	Right panel tab listing all comments per slide
Resolve Comments	Mark comments as resolved
Delete Comments	Remove individual comments

Export Options

Format	Description
PPTX	PowerPoint with slide images + editable text overlay layers
PDF	Multi-page PDF of all slides
PNG ZIP	Download all slides as a ZIP of PNG images
Animated GIF	Export slides as an animated GIF slideshow (custom duration)
Video	Clean MP4 with logo removed and audio preserved

Presentation Mode

Feature	Description
Fullscreen	`F5` enters distraction-free fullscreen presentation
Arrow Navigation	Left/Right arrows to navigate slides
Escape to Exit	Press Escape to return to editor

UI / UX

Feature	Description
Glassmorphism Theme	Modern frosted glass dark theme with Inter font
Organized Toolbar	Labeled tool groups: TOOLS, EDIT, INSERT, SLIDE, MANAGE
Drag & Drop Upload	Drop PPTX files anywhere to upload
Auto-Save Indicator	Pulsing orange dot when unsaved, green checkmark on save
Loading Overlay	Spinner during long operations
Toast Notifications	Success / error / info messages with icons
Crossfade Navigation	Smooth slide transitions
Responsive Toolbar	Wraps on smaller screens

Installation

Fastest path — see INSTALL.md for the full cross-platform guide. TL;DR:

# Clone the repo from GitHub to your machine
git clone https://github.com/Layruss98266/slidecraft.git

# Move into the project folder
cd slidecraft

# Run the launcher for your OS — it will:
#   1. Create a Python virtual environment in .venv/
#   2. Install everything listed in requirements.txt
#   3. Warn you if LibreOffice isn't installed
#   4. Start the Flask server on http://127.0.0.1:5050
./run.sh              # macOS / Linux  (chmod +x run.sh once if needed)
# or run.bat          # Windows (cmd.exe / double-click)
# or .\run.ps1        # Windows (PowerShell)

The launcher creates .venv/, installs requirements.txt, warns if LibreOffice isn't installed, then starts the server at http://127.0.0.1:5050. First run takes ~3 min (LibreOffice optional but strongly recommended for high-fidelity slide rendering).

How to Run

Option A — one-command launcher (recommended)

After the install steps above, just re-run the launcher whenever you want to start the editor. It's idempotent — it skips the venv/install steps if they've already been done.

# macOS / Linux
./run.sh

:: Windows (cmd.exe)
run.bat

# Windows (PowerShell)
.\run.ps1

Then open http://127.0.0.1:5050 in any modern browser (Chrome, Edge, Firefox, Safari).

Option B — manual (if you already have the venv set up)

# 1. Activate the virtual environment
source .venv/bin/activate        # macOS / Linux
.venv\Scripts\Activate.ps1       # Windows PowerShell
.venv\Scripts\activate.bat       # Windows cmd.exe

# 2. Start the server (Ctrl+C to stop)
python app.py

Configuration via environment variables

All optional — sensible defaults if unset.

# Bind on all interfaces so other machines on your LAN can reach the editor.
# WARNING: there's no built-in auth — only do this on a trusted network.
HOST=0.0.0.0 python app.py

# Use a different port (e.g. if 5050 is taken)
PORT=8080 python app.py

# Allow larger uploads (default is 60 MB total per request)
MAX_UPLOAD_MB=200 python app.py

# Combine multiple
HOST=0.0.0.0 PORT=8080 MAX_UPLOAD_MB=200 python app.py

On Windows PowerShell use $env:HOST = "0.0.0.0"; python app.py instead.

See INSTALL.md for the full list.

Stopping the server

Press Ctrl+C in the terminal where it's running. State on disk (uploaded PPTX, slide images, history snapshots) is preserved — the next start picks up where you left off.

Typical first-run flow

1. Run the launcher                        →  server listens on :5050
2. Open http://127.0.0.1:5050 in a browser →  empty editor
3. Click "Upload" and pick a .pptx file    →  slides appear in the strip
4. Edit / OCR / watermark / filter / etc.  →  every change auto-saves
5. Click "Export" → "PowerPoint (PPTX)"    →  download the edited deck

The first slide upload triggers a one-time ~64 MB EasyOCR model download to ~/.EasyOCR/. After that, OCR is instant.

Installation (Step by Step)

Step 1: Install Python

SlideCraft requires Python 3.10 or higher.

Windows

Go to python.org/downloads
Download the latest Python installer (e.g. Python 3.12.x)
Run the installer
IMPORTANT: Check the box "Add Python to PATH" at the bottom of the installer
Click "Install Now"
Verify installation — open Command Prompt and run:
```
python --version
```
You should see something like Python 3.12.x

macOS

Install Homebrew if you don't have it:

/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"

Install Python:
```
brew install python
```
Verify:
```
python3 --version
```

Linux (Ubuntu/Debian)

sudo apt update
sudo apt install python3 python3-pip python3-venv
python3 --version

Step 2: Download SlideCraft

Option A — Clone with Git (recommended):

git clone https://github.com/YOUR_USERNAME/slidecraft.git
cd slidecraft

Option B — Download ZIP:

Click the green "Code" button on GitHub
Click "Download ZIP"
Extract the ZIP file
Open a terminal/command prompt in the extracted folder

Step 3: Create a Virtual Environment (Recommended)

This keeps SlideCraft's packages separate from your system Python.

Windows:

python -m venv venv
venv\Scripts\activate

macOS / Linux:

python3 -m venv venv
source venv/bin/activate

You'll see (venv) at the start of your terminal prompt when the virtual environment is active.

Step 4: Install Dependencies

pip install -r requirements.txt

This installs all required packages:

Package	What it does	Size
`flask`	Web server	~2 MB
`python-pptx`	Read/write PowerPoint files	~5 MB
`Pillow`	Image processing & filters	~10 MB
`opencv-python-headless`	Inpainting, video frames, filters	~40 MB
`easyocr`	Text detection (OCR)	~150 MB
`moviepy`	Video audio handling & trimming	~5 MB
`numpy`	Array operations	(with opencv)
`torch`	AI backend for EasyOCR	~200-800 MB

Note: First install may take 2-5 minutes. The torch package is the largest (~200MB CPU-only).

If you get an error with pip, try:

python -m pip install -r requirements.txt

Optional — for QR code generation:

pip install qrcode[pil]

Step 5: Run SlideCraft

Windows:

set PYTHONIOENCODING=utf-8 && python app.py

macOS / Linux:

PYTHONIOENCODING=utf-8 python3 app.py

Dev mode: To enable Flask debug mode (auto-reload on code changes), set FLASK_DEBUG=true:
FLASK_DEBUG=true PYTHONIOENCODING=utf-8 python app.py

You should see:

 * Running on http://0.0.0.0:5050

Step 6: Open in Browser

Open your browser and go to:

http://localhost:5050

That's it! SlideCraft is running.

First time using OCR? The first time you click the "OCR" button, EasyOCR will download its text detection models (~100 MB). This is a one-time download.

Optional: Install LibreOffice (for better PPTX conversion)

LibreOffice enables high-fidelity PPTX-to-image conversion. Without it, SlideCraft falls back to extracting embedded images.

Windows: Download from libreoffice.org and install.

macOS:

brew install --cask libreoffice

Linux:

sudo apt install libreoffice

Usage Guide

Upload a Presentation

Click Upload in the header, or drag & drop a .pptx file anywhere
Slides are converted to images and loaded into the editor

Remove NotebookLM Logo (Slides)

Click Remove Logo in the header — all slides processed via OpenCV inpainting

Edit Text with OCR

Click OCR in the toolbar — text regions highlighted with blue dashed boxes
Click any region — creates a background cover + editable text overlay
Edit text in the Properties panel, adjust font/size/color/weight
Click Apply to burn changes permanently into the slide image

Manual Text Replacement (Cover & Type)

Press C or click the Cover tool
Draw a rectangle over text to replace
Type replacement text in Properties panel
Click Apply to bake

Draw Shapes & Freehand

Use toolbar tools: Rectangle (R), Circle, Line, Callout, Freehand Draw
All shapes are resizable and repositionable
Right-click for context menu: Duplicate, Delete, Bring to Front/Back

Apply Image Filters

Switch to Filters tab in the right panel
Click any filter button — applies instantly to the current slide

Manage Templates

Click Templates icon in toolbar
Save current state as a named template
Load any saved template to restore slides + overlays

Version History

Click History icon in toolbar
Save Version to snapshot current state
Restore any previous version

Comments

Click Comments tool, then click on the slide to place a pin
Enter comment text — visible as numbered pins on the canvas
Switch to Comments tab in the right panel to view/resolve/delete

Export

Button	Output
PPTX	PowerPoint with overlays as editable layers
PDF	Multi-page PDF
PNG	ZIP archive of all slides as PNG
GIF	Animated GIF slideshow
Apply	Burns overlays into slide images (no layers)

Remove Logo from Video

Click Video in the header (or go to /video)
Upload video file (any format)
Drag the red box over the logo on the preview frame
Click Remove Logo — watch real-time progress
Click Stop to cancel if needed
Download the clean video when done

Keyboard Shortcuts

Press ? in the editor to see the full cheatsheet.

Key	Action
`S`	Select tool
`T`	Text tool
`R`	Rectangle tool
`C`	Cover & Type tool
`G`	Toggle snap grid
`?`	Show all keyboard shortcuts
`F5`	Fullscreen presentation mode
`Ctrl+Z`	Undo
`Ctrl+Y`	Redo
`Ctrl+C`	Copy selected overlay
`Ctrl+V`	Paste overlay
`Ctrl+S`	Save current slide
`Delete`	Delete selected overlay
`Escape`	Exit presentation / close modals
`Arrow Left/Right`	Navigate slides

Project Structure

slidecraft/
├── app.py                  # Flask backend — 44 API routes
├── requirements.txt        # Python dependencies
├── LICENSE                 # MIT license
├── .gitignore              # Git ignore rules
├── slide_data.json         # Auto-generated overlay data store
├── comments.json           # Auto-generated comments store
├── static/
│   └── slides/             # Slide JPG images (auto-generated)
├── uploads/                # Uploaded PPTX files
├── exports/                # Exported PPTX/PDF/ZIP/GIF files
├── videos/                 # Uploaded & processed videos
├── history/                # Version history snapshots
├── templates_saved/        # Saved slide templates
└── templates/
    ├── index.html          # Main slide editor UI (~3900 lines)
    └── video.html          # Video logo remover UI

API Reference (44 Routes)

Slide Endpoints

Method	Endpoint	Description
`GET`	`/`	Main editor page
`GET`	`/api/slide/<num>`	Get overlays + notes for a slide
`POST`	`/api/slide/<num>`	Save overlays + notes
`POST`	`/api/slide/<num>/bake`	Burn overlays into slide image
`POST`	`/api/slide/<num>/preview`	Render composite preview with overlays
`POST`	`/api/slide/<num>/filter`	Apply image filter
`POST`	`/api/slide/<num>/crop`	Crop slide image
`POST`	`/api/slide/<num>/rotate`	Rotate slide image
`POST`	`/api/upload`	Upload PPTX file
`POST`	`/api/remove-logo`	Remove NotebookLM logo from all slides
`POST`	`/api/reorder`	Reorder slides
`POST`	`/api/upload-image`	Upload image for overlay use
`POST`	`/api/ocr/<num>`	Run EasyOCR on a slide
`POST`	`/api/sample-color/<num>`	Sample bg color, text color, font weight
`POST`	`/api/qr-generate`	Generate QR code image
`POST`	`/api/watermark`	Add text watermark to all slides
`POST`	`/api/watermark-image`	Add image/logo watermark to all slides
`POST`	`/api/detect-watermark/<num>`	Auto-detect watermark regions in a slide
`POST`	`/api/remove-watermark/<num>`	Remove watermark from single slide
`POST`	`/api/remove-watermark-all`	Remove watermark from all slides
`POST`	`/api/find-replace`	Find and replace text in overlays

Export Endpoints

Method	Endpoint	Description
`POST`	`/api/export`	Export as PPTX
`POST`	`/api/export-pdf`	Export as PDF
`POST`	`/api/export-png-zip`	Export as PNG ZIP
`POST`	`/api/export-gif`	Export as animated GIF

Batch Endpoints

Method	Endpoint	Description
`POST`	`/api/batch/remove-logo`	Bulk upload up to 20 PPTX, remove logos, return ZIP

Template Endpoints

Method	Endpoint	Description
`GET`	`/api/templates`	List saved templates
`POST`	`/api/templates/save`	Save current state as template
`POST`	`/api/templates/load`	Load a template
`POST`	`/api/templates/delete`	Delete a template

Version History Endpoints

Method	Endpoint	Description
`GET`	`/api/history`	List all versions
`POST`	`/api/history/save`	Save a version snapshot
`POST`	`/api/history/restore`	Restore a version

Comment Endpoints

Method	Endpoint	Description
`GET`	`/api/comments/<num>`	Get comments for a slide
`POST`	`/api/comments/<num>`	Add a comment
`POST`	`/api/comments/<num>/resolve`	Resolve a comment
`POST`	`/api/comments/<num>/delete`	Delete a comment

Video Endpoints

Method	Endpoint	Description
`GET`	`/video`	Video logo remover page
`POST`	`/api/video/upload`	Upload video file
`POST`	`/api/video/preview-frame`	Get a preview frame
`POST`	`/api/video/remove-logo`	Start logo removal (returns jobId)
`GET`	`/api/video/progress/<jobId>`	Poll job progress
`POST`	`/api/video/stop/<jobId>`	Stop a running job
`GET`	`/api/video/download/<file>`	Download processed video

Tech Stack

Layer	Technology
Backend	Python 3.10+, Flask 3.x
PPTX	python-pptx
Image Processing	Pillow, OpenCV (inpainting, filters)
OCR	EasyOCR (PyTorch-based, no external binary)
Video	OpenCV (frames), MoviePy (audio mux, trim)
Frontend	Vanilla JavaScript, HTML5 Canvas, CSS3
Theme	Glassmorphism dark theme, Inter font
Icons	Inline SVG stroke icons

How It Works

Logo Removal

Uses OpenCV's Telea inpainting algorithm:

Binary mask marks the logo region
Algorithm propagates pixels from all surrounding edges inward
Follows natural color gradients and textures
Result: seamless removal with no flat bands

OCR Text Editing

EasyOCR detects text regions with bounding boxes
sample-color API analyzes the region:
- Background color from a strip above the text
- Text color from pixels with highest contrast
- Font weight from ink density (normal/semibold/bold/extrabold)
Cover rectangle + text overlay created with matched colors/weight
Apply burns it into the image using Pillow with correct system font

Video Processing

Background thread processes each frame with OpenCV inpainting
Progress reported via polling API (frame count, percentage, ETA)
Cancellation supported — stops the thread cleanly
MoviePy muxes audio from original into the clean video

Troubleshooting

Issue	Solution
OCR not working	EasyOCR downloads models on first use (~100MB). Ensure internet access.
`UnicodeEncodeError` on Windows	Run with `set PYTHONIOENCODING=utf-8 && python app.py`
Video processing slow	~10-30 fps depending on resolution. 1080p 5min = ~5-8 min.
Fonts don't match exactly	Use the Font Family dropdown in Properties.
PPTX upload shows blank slides	Install LibreOffice for proper conversion.
QR code shows placeholder	Install `qrcode` package: `pip install qrcode[pil]`
Stop button not working	Wait a few seconds — cancellation is checked every frame.

Contributing

Fork the repo
Create a feature branch (git checkout -b feature/amazing-feature)
Commit your changes (git commit -m 'Add amazing feature')
Push to the branch (git push origin feature/amazing-feature)
Open a Pull Request

License

MIT License. See LICENSE for details.

Built with Flask + Pillow + OpenCV + EasyOCR + MoviePy
44 API routes | 45+ features | 0 build steps

Name		Name	Last commit message	Last commit date
Latest commit History 19 Commits
.github		.github
history/20260409_232425		history/20260409_232425
static		static
templates		templates
tests		tests
.gitignore		.gitignore
INSTALL.md		INSTALL.md
LICENSE		LICENSE
README.md		README.md
app.py		app.py
pyproject.toml		pyproject.toml
requirements-dev.txt		requirements-dev.txt
requirements.txt		requirements.txt
run.bat		run.bat
run.ps1		run.ps1
run.sh		run.sh
run_tests.py		run_tests.py

Uh oh!

Folders and files

Latest commit

History

Repository files navigation

SlideCraft

What is SlideCraft?

Screenshots

Features at a Glance (40+)

Drawing & Shape Tools

Text Editing & OCR

Typography Controls

Image Filters

Logo & Watermark Removal (Slides)

Custom Watermark (Add Your Own)

Video Logo Remover

Editing Tools

Insert Tools

Templates & Version History

Comments & Annotations

Export Options

Presentation Mode

UI / UX

Installation

How to Run

Option A — one-command launcher (recommended)

Option B — manual (if you already have the venv set up)

Configuration via environment variables

Stopping the server

Typical first-run flow

Installation (Step by Step)

Step 1: Install Python

Step 2: Download SlideCraft

Step 3: Create a Virtual Environment (Recommended)

Step 4: Install Dependencies

Step 5: Run SlideCraft

Step 6: Open in Browser

Optional: Install LibreOffice (for better PPTX conversion)

Usage Guide

Upload a Presentation

Remove NotebookLM Logo (Slides)

Edit Text with OCR

Manual Text Replacement (Cover & Type)

Draw Shapes & Freehand

Apply Image Filters

Manage Templates

Version History

Comments

Export

Remove Logo from Video

Keyboard Shortcuts

Project Structure

API Reference (44 Routes)

Slide Endpoints

Export Endpoints

Batch Endpoints

Template Endpoints

Version History Endpoints

Comment Endpoints

Video Endpoints

Tech Stack

How It Works

Logo Removal

OCR Text Editing

Video Processing

Troubleshooting

Contributing

License

About

Resources

License

Uh oh!

Stars

Watchers

Forks

Releases

Sponsor this project

Uh oh!

Packages 0

Uh oh!

Packages