Skip to content

USAGE.md

Latest commit

 

History

History
205 lines (148 loc) · 6.25 KB

USAGE.md

File metadata and controls

205 lines (148 loc) · 6.25 KB

Usage Guide

A comprehensive guide on how to use Sharpify GUI - a PyQt6-based desktop app to upscale images and videos using Real-ESRGAN.

Table of Contents

Overview

Sharpify GUI is a professional-grade tool designed to upscale images and videos efficiently. Built with PyQt6, it provides an intuitive desktop interface for:

  • Image Upscaling: Upscale JPG, PNG, BMP, TIFF, and WebP images by 4x.
  • Video Upscaling: Upscale MP4, AVI, MKV, MOV, WMV, and FLV videos by 4x.
  • Batch Processing: Enqueue multiple files for sequential processing with progress tracking.
  • Model Selection: Choose between different Real-ESRGAN models optimized for various content types.
  • GPU Acceleration: Leverage CUDA/OpenCL for significantly faster upscaling.
  • Advanced Configuration: Fine-tune settings like tile size, output format, FPS, and quality.
  • Comprehensive Logging: Keep track of the upscaling process with detailed, timestamped logs.

Quick Start

Prerequisites

  • Windows Operating System
  • A modern GPU with support for Vulkan.
  • Git (for cloning the repository)

Installation

Clone from GitHub

# Clone the repository
git clone https://github.com/uikraft-hub/sharpify-gui.git

# Navigate to the project directory
cd sharpify-gui

# Install required dependencies
pip install -r requirements.txt

Running the Application

  1. Navigate to the src directory: 
    cd src
  2. Run the application: 
    python main.py

The application window will open, and you can start using it.

User Interface Guide

Main Interface Components

1. Left Panel

  • Input Files: Add files and folders to the processing queue. You can also drag and drop files here.
  • Output Settings: Select the folder where the upscaled files will be saved.
  • Quick Settings: Quickly select the upscaling model.
  • Controls: Start/stop the upscaling process and access advanced settings.

2. Right Panel

  • Progress: View the progress of the current file and the overall batch.
  • Processing Log: See detailed, timestamped logs of the upscaling process.

Supported File Types

Image Formats

  • JPG
  • JPEG
  • PNG
  • BMP
  • TIFF
  • WebP

Video Formats

  • MP4
  • AVI
  • MKV
  • MOV
  • WMV
  • FLV

Upscaling Options

Model Selection

You can choose from the following Real-ESRGAN models:

  • Anime Image/Video 4x: Optimized for upscaling anime-style images and videos.
  • General Image/Video 4x: A general-purpose model for all other types of images and videos.
  • Anime Photos 4x: A model specifically for anime-style photos.

Output Format

For images, you can choose the output format:

  • JPG: Good for photos, smaller file size.
  • PNG: Lossless format, good for graphics.
  • WebP: Modern format with good compression.

Videos are output in the same format as the input file.

Advanced Settings

You can access the advanced settings by clicking the "Advanced Settings" button.

AI Model Settings

  • Model: Select the Real-ESRGAN model to use.

Performance Settings

  • Use GPU Acceleration: Enable or disable GPU usage.
  • Tile Size: Controls GPU memory usage. Lower values use less memory but are slower.

Video Processing Settings

  • Output FPS: Set the frames per second for the output video.
  • Video Quality (CRF): Control the quality of the output video. Lower values mean higher quality and larger file sizes.

Output Format Settings

  • Image Format: Choose the output format for upscaled images.

Troubleshooting

Common Issues and Solutions

Application does not start

Possible Causes:

  • Missing dependencies.
  • Python is not installed correctly.

Solutions:

  1. Make sure you have installed all the dependencies from requirements.txt.
  2. Ensure you are using a supported version of Python.

Upscaling fails

Possible Causes:

  • The input file is corrupted.
  • The selected model is not compatible with the input file.
  • Insufficient GPU memory.

Solutions:

  1. Try a different input file.
  2. Try a different upscaling model.
  3. If using GPU, try reducing the tile size in the advanced settings.

Best Practices

Performance Optimization

  • Use GPU acceleration for the best performance.
  • Adjust the tile size based on your GPU's memory.
  • Process files in batches to save time.

File Management

  • Organize your input and output files in separate folders.
  • Use the "Clear All" button to clear the file queue before starting a new batch.

API Reference

The application is not designed to be used as a library, but the core logic is contained in the following classes:

Core Classes

app.main_window.AnimeUpscalerGUI

The main application window.

app.workers.UpscaleWorker

Handles the upscaling process in a separate thread.

FAQ

General Questions

Q: Is this application free to use? A: Yes, this is an open-source project under the MIT license.

Q: Do I need a powerful computer to run this application? A: A modern GPU is recommended for the best performance, but the application can also run on the CPU.

Technical Questions

Q: Can I use my own Real-ESRGAN models? A: Currently, you can only use the models included with the application.

Q: Where are the upscaled files saved? A: You can select the output folder using the "Select Output Folder" button.

📞 Support