GPU-Accelerated Video Filter Pipeline

A real-time video processing application demonstrating GPU acceleration using CUDA. Implements various image filters on both CPU and GPU with performance comparisons.

Features

• Real-time video playback with FFmpeg
• CPU and GPU implementations of filters:
  • Grayscale conversion
  • Brightness/Contrast adjustment
  • Box blur
• Performance metrics and FPS display
• Side-by-side CPU vs GPU performance comparison

Requirements

Hardware

• NVIDIA GPU with compute capability 5.0 or higher
• Minimum 4GB GPU memory

Software

• CMake 3.18 or later
• C++17 compatible compiler
• CUDA Toolkit 11.0 or later
• FFmpeg development libraries (avformat, avcodec, avutil, swscale)
• SDL2 development library

Building

Windows

1. Install CUDA Toolkit from NVIDIA's website
2. Install Visual Studio with C++ support
3. Download FFmpeg development libraries and extract to C:\ffmpeg
4. Install SDL2 development libraries

mkdir build
cd build
cmake .. -DFFMPEG_INCLUDE_DIR="C:/ffmpeg/include" -DFFMPEG_LIB_DIR="C:/ffmpeg/lib"
cmake --build . --config Release

Linux

1. Install dependencies:

sudo apt-get install cmake build-essential
sudo apt-get install libavformat-dev libavcodec-dev libavutil-dev libswscale-dev
sudo apt-get install libsdl2-dev
# Install CUDA Toolkit from NVIDIA

2. Build:

mkdir build
cd build
cmake ..
make -j$(nproc)

Usage

./video_filter <path_to_video_file>

Controls

• 1, 2, 3: Switch between filters
  • 1: Grayscale
  • 2: Brightness/Contrast
  • 3: Box Blur
• 0: No filter (passthrough)
• M: Toggle between CPU and GPU processing
• Space: Pause/Resume
• Q or ESC: Quit

Performance

Expected performance on typical hardware (GTX 1660 Ti, 1080p video):

• CPU: ~10-15 FPS
• GPU: ~150-200 FPS
• Speedup: 10-15x

Project Structure

src/
├── main.cpp                    # Application entry point
├── video/
│   ├── Frame.h                 # Frame data structure
│   └── VideoReader.h/cpp       # FFmpeg video decoder
├── display/
│   ├── Display.h/cpp           # SDL2 rendering
│   └── PerformanceOverlay.h/cpp # FPS overlay
├── filters/
│   ├── Filter.h                # Filter interface
│   ├── CPUFilters.h/cpp        # CPU filter implementations
│   └── GPUFilters.h/cu         # CUDA kernel implementations
└── gpu/
    └── GPUManager.h/cpp        # CUDA memory management

License

Educational project for learning GPU programming and video processing.