High-performance batch image converter with GPU acceleration
Desktop.2025.12.03.-.07.28.15.01.mp4
- Features
- Screenshots
- System Requirements
- Installation
- Building from Source
- Usage Guide
- Technical Details
- Troubleshooting
- Contributing
- License
- Batch Processing - Convert hundreds of images in seconds
- GPU Acceleration - CUDA and OpenCL support for 5-10x faster processing
- Multiple Formats - JPG, PNG, WEBP, TIF, BMP, GIF
- Drag & Drop - Simple, intuitive interface
- Recursive Scanning - Automatically processes entire folder structures
- Smart Resize - Scale images by percentage (10-100%)
- Color Conversion - Auto-detect, BGR, Grayscale, RGBA modes
- Denoise Filter - Advanced non-local means denoising with adjustable strength
- Quality Control - Adjustable compression quality (10-100%)
- Multi-threaded - Utilizes all CPU cores
- Mixed Pipeline - Intelligently switches between GPU/CPU based on task
- Real-time Progress - Live conversion status with animated progress bar
- Smart Output - Automatic file naming prevents overwrites
[Main Interface]
┌─────────────────────────────────────────────────────┐
│ ⚡ QuickConvert 📁 ✕ │
│ High-Quality • GPU Detected • Batch Processing │
├─────────────────────────────────────────────────────┤
│ │
│ Quality: 95% ████████████████░░ JPG PNG WEBP │
│ TIF BMP GIF │
│ │
│ ┌─────────────────────┐ ┌───────────────────┐ │
│ │ Drop Images or │ │ Settings Panel │ │
│ │ Folders Here │ │ ☑ Resize: 100% │ │
│ │ │ │ ☐ Color Convert │ │
│ │ [Progress Bar] │ │ ☐ Denoise │ │
│ └─────────────────────┘ └───────────────────┘ │
│ │
│ Output: C:\Converted GPU: ON │
└─────────────────────────────────────────────────────┘
- OS: Windows 10 (64-bit) or later
- CPU: Dual-core processor
- RAM: 2 GB
- Disk: 100 MB free space
- OS: Windows 11 (64-bit)
- CPU: Quad-core processor or better
- RAM: 8 GB or more
- GPU: NVIDIA GPU with CUDA support (GTX 900 series or newer)
- OR AMD/Intel GPU with OpenCL support
- NVIDIA: GTX 900+ series, RTX series (CUDA)
- AMD: Radeon RX 400+ series (OpenCL)
- Intel: Iris Xe, Arc series (OpenCL)
- Download the latest release from https://github.com/codeboxqc/QuickPic/releases/tag/1
- Extract
QuickConvert.zipto your desired location - Run
QuickConvert.exe - (Optional) Create a desktop shortcut
No installation required! Just extract and run.
-
Visual Studio 2022 (Community Edition or higher)
- Download: https://visualstudio.microsoft.com/
- Workloads required:
- "Desktop development with C++"
- Windows 10 SDK (10.0.19041.0 or later)
-
OpenCV 4.13.0 with CUDA support
- Download pre-built: https://opencv.org/releases/
- OR build from source with CUDA enabled (see below)
-
CUDA Toolkit 12.x (for GPU acceleration)
- Download: https://developer.nvidia.com/cuda-downloads
- Required even if you only use OpenCL (for headers)
- CMake 3.20+ (if building OpenCV from source)
- Git for cloning the repository
- Download Visual Studio 2022 Community
- Run installer and select:
- ✅ Desktop development with C++
- ✅ C++ CMake tools for Windows
- ✅ Windows 10 SDK (latest version)
- Complete installation (requires ~7 GB)
- Download CUDA Toolkit 12.x from NVIDIA
- Run installer with default settings
- Verify installation:
Should output:
nvcc --version
Cuda compilation tools, release 12.x
- Download OpenCV 4.13.0 from https://opencv.org/releases/
- Extract to
C:\opencv - Your directory structure should be:
C:\opencv\ ├── build\ │ ├── x64\ │ │ └── vc16\ │ │ ├── bin\ │ │ │ └── opencv_world4130.dll │ │ └── lib\ │ │ └── opencv_world4130.lib │ └── include\ │ └── opencv2\ └── sources\
:: Create the master directory
mkdir E:\opencv
cd /d E:\opencv
:: Clone the main OpenCV repository (Branch 4.x or specifically tag 4.13.0)
git clone --branch 4.13.0 https://github.com/opencv/opencv.git opencv-src
:: Clone the OpenCV Contrib repository (Must exactly match the main version)
git clone --branch 4.13.0 https://github.com/opencv/opencv_contrib.git opencv_contrib
:: Create the folder where the compiled binaries will go
mkdir build
cd /d E:\opencv\build
cd /d E:\opencv\build
del /q CMakeCache.txt
rmdir /s /q CMakeFiles
cmake -G "Visual Studio 17 2022" -A x64 ^
-D CMAKE_VS_GLOBALS=VcpkgEnabled=false ^
-D CMAKE_BUILD_TYPE=Release ^
-D CMAKE_INSTALL_PREFIX=E:/opencv/install ^
-D CMAKE_CXX_STANDARD=17 ^
-D OPENCV_EXTRA_MODULES_PATH=E:/opencv/opencv_contrib/modules ^
-D BUILD_opencv_world=ON ^
-D WITH_CUDA=ON ^
-D WITH_CUDNN=ON ^
-D OPENCV_DNN_CUDA=ON ^
-D CUDA_FAST_MATH=ON ^
-D WITH_OPENVINO=ON ^
-D OPENCV_DNN_OPENVINO=ON ^
-D WITH_TBB=ON ^
-D TBB_DIR=E:/opencv/tbb/lib/cmake/TBB ^
-D WITH_EIGEN=ON ^
-D EIGEN_INCLUDE_PATH=E:/opencv/eigen ^
-D WITH_JPEG=ON ^
-D WITH_PNG=ON ^
-D WITH_TIFF=ON ^
-D WITH_WEBP=ON ^
-D WITH_OPENEXR=ON ^
-D BUILD_ZLIB=ON ^
-D BUILD_SHARED_LIBS=ON ^
E:/opencv/opencv-src
1. Install Eigen via Git
Eigen is open-source, so we can clone it directly into the correct folder using Git.
DOS
cd /d E:\opencv
:: Clone the stable 3.4.0 release directly into E:\opencv\eigen
git clone --branch 3.4.0 https://gitlab.com/libeigen/eigen.git E:\opencv\eigen
2. Install Intel TBB via cURL
Because we want the pre-compiled Windows binaries for TBB (so you don't have to spend hours compiling TBB itself), we will use curl to download the zip file from GitHub, and tar to extract it.
DOS
cd /d E:\opencv
:: Download the TBB release zip file
curl -L -o tbb.zip https://github.com/uxlfoundation/oneTBB/releases/download/v2021.12.0/oneapi-tbb-2021.12.0-win.zip
:: Extract the zip file
tar -xf tbb.zip
:: Rename the extracted folder to just "tbb" so CMake finds it easily
rename oneapi-tbb-2021.12.0 tbb
:: Clean up the zip file
del tbb.zip
3. What about OpenVINO?
You can clone OpenVINO using Git (git clone --recurse-submodules https://github.com/openvinotoolkit/openvino.git), BUT compiling OpenVINO from source on Windows takes an incredibly long time and requires dozens of extra dependencies (Python, Ninja, etc.).
For OpenVINO, it is highly recommended to download the pre-built Windows Archive. You can do that via the command line too:
DOS
cd /d E:\opencv
:: Download the pre-built OpenVINO toolkit archive for Windows
curl -L -o openvino.zip https://storage.openvinotoolkit.org/repositories/openvino/packages/2024.1/windows/w_openvino_toolkit_windows_2024.1.0.15008.f4afc983258_x86_64.zip
:: Extract it
tar -xf openvino.zip
:: Rename it to "openvino" for simplicity
rename w_openvino_toolkit_windows_2024.1.0.15008.f4afc983258_x86_64 openvino
:: Clean up the zip
del openvino.zip
Verify Your Setup
Once those commands finish, type dir in your command prompt. Your E:\opencv folder should now look exactly like this:
Plaintext
E:\opencv> dir
...
<DIR> build
<DIR> eigen
<DIR> opencv-src
<DIR> opencv_contrib
<DIR> openvino
<DIR> tbb
Phase 5: Compile Your Ultimate OpenCV Build
Your configuration blueprint is ready. Run these two commands to compile and finalize your bare-metal installation:
Compile the source files (This will utilize 8 CPU threads to expedite the process):
DOS
cmake --build . --config Release --parallel 8
(This phase will take roughly 30 to 60 minutes as it builds all the custom AI and CUDA modules).
Install and organize the binaries:
DOS
cmake --install . --config Release
Step 1: Open the file instantly
Open your command prompt and run this exact command. It will open the hidden file directly in Notepad:
DOS
notepad E:\opencv\opencv_contrib\modules\cudev\include\opencv2\cudev\ptr2d\zip.hpp
Step 2: Change the two broken words
In Notepad, scroll down to the very bottom of the file (around line 182).
Look for this exact text:
C++
_LIBCUDACXX_BEGIN_NAMESPACE_STD
Change it to:
C++
_CCCL_BEGIN_NAMESPACE_STD
Then, look a few lines below that (around line 201) for this exact text:
C++
_LIBCUDACXX_END_NAMESPACE_STD
Change it to:
C++
_CCCL_END_NAMESPACE_STD
Step 3: Save and Resume!
Save the file in Notepad (Ctrl + S) and close Notepad.
Go straight back to your Visual Studio Command Prompt. You do not need to clear the cache or run CMake again.
Just run the build command to resume exactly where it failed:
DOS
cmake --build . --config Release --parallel 8-
Open System Properties → Advanced → Environment Variables
-
Add to System Variables:
- Variable:
OPENCV_DIR - Value:
C:\opencv\build
- Variable:
-
Add to Path variable:
C:\opencv\build\x64\vc16\binC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.x\bin
-
Click OK and restart any open command prompts
git clone https://github.com/yourusername/quickconvert.git
cd quickconvert-
Open
Quick.slnin Visual Studio 2022 -
Right-click project → Properties
-
C/C++ → General → Additional Include Directories:
C:\opencv\build\include $(CUDA_PATH)\include -
Linker → General → Additional Library Directories:
C:\opencv\build\x64\vc16\lib $(CUDA_PATH)\lib\x64 -
Linker → Input → Additional Dependencies:
opencv_world4130.lib cudart.lib -
Build Events → Post-Build Event (auto-copy DLLs):
xcopy /Y /D "C:\opencv\build\x64\vc16\bin\opencv_world4130.dll" "$(OutDir)" xcopy /Y /D "C:\opencv\build\x64\vc16\bin\opencv_videoio_ffmpeg4130_64.dll" "$(OutDir)"
- Select Release configuration and x64 platform
- Build → Rebuild Solution (Ctrl+Shift+B)
- Check for errors in Output window
- Executable will be in:
x64\Release\QuickConvert.exe
-
Copy required DLLs to executable directory:
copy "C:\opencv\build\x64\vc16\bin\*.dll" "x64\Release\"
-
Run
QuickConvert.exe -
Check if GPU is detected (shown in subtitle)
-
Drop a test image to verify conversion works
- Launch QuickConvert
- Set Output Folder (click 📁 icon)
- Adjust Quality (slider, 10-100%)
- Select Format (JPG, PNG, WEBP, etc.)
- Drag & Drop images or folders
- Wait for completion (progress bar + popup result)
- ☑ Enable "Resize" checkbox
- Enter percentage (10-100%)
- Example:
50= half size,200= not allowed (max 100%)
- ☑ Enable "Convert Color" checkbox
- Select mode:
- Auto - Keep original format
- BGR - Standard RGB (3 channels)
- **Gray