Library of programming functions mainly aimed at real-time vision

OpenCV

Browse by Company

Adobe, Apowersoft, Ashampoo, Autodesk, Avast, Corel, Cyberlink, Google, iMyFone, iTop, Movavi, PassFab, Passper, Stardock, Tenorshare, Wargaming, Wondershare

OpenCV

  -  174 MB  -  Open Source
Free Download

Security Status

  • Latest Version

    OpenCV 4.10.0 LATEST

  • Review by

    Daniel Leblanc

  • Operating System

    Windows XP / Vista / Windows 7 / Windows 8 / Windows 10

  • User Rating

    Click to vote

  • Author / Product

    OpenCV Team / External Link

  • Filename

    opencv-4.10.0-windows.exe

OpenCV (Open Source Computer Vision Library) is released under a BSD license and hence it’s free for both academic and commercial use. It has C++, Python, and Java interfaces and supports Windows, Linux, macOS, iOS, and Android. It was designed for computational efficiency and with a strong focus on real-time applications. Written in optimized C/C++, the library can take advantage of multi-core processing. Enabled with OpenCL, it can take advantage of the hardware acceleration of the underlying heterogeneous compute platform.

Adopted all around the world, OpenCV has more than 47 thousand people of the user community and an estimated number of downloads exceeding 14 million. Usage ranges from interactive art to mines inspection, stitching maps on the web, or through advanced robotics.

OpenCV (Open Source Computer Vision Library) is an open-source computer vision and machine learning software library. It was built to provide a common infrastructure for computer vision applications and to accelerate the use of machine perception in commercial products. Being a BSD-licensed product, It makes it easy for businesses to utilize and modify the code.

The library has more than 2500 optimized algorithms, which includes a comprehensive set of both classic and state-of-the-art computer vision and machine learning algorithms. These algorithms can be used to detect and recognize faces, identify objects, classify human actions in videos, track camera movements, track moving objects, extract 3D models of objects, produce 3D point clouds from stereo cameras, stitch images together to produce a high-resolution image of an entire scene, find similar images from an image database, remove red eyes from images taken using flash, follow eye movements, recognize scenery and establish markers to overlay it with augmented reality, etc.

It has more than 47 thousand people of the user community and an estimated number of downloads exceeding 14 million. The library is used extensively in companies, research groups, and governmental bodies.

Open CV has C++, Python, Java, and MATLAB interfaces and supports Windows, Linux, Android, and macOS. It leans mostly towards real-time vision applications and takes advantage of MMX and SSE instructions when available. A full-featured CUDA and OpenCL interfaces are being actively developed right now. There are over 500 algorithms and about 10 times as many functions that compose or support those algorithms. Open CV is written natively in C++ and has a templated interface that works seamlessly with STL containers.

Also Available: Download OpenCV for Mac

Download OpenCV Latest Version

  • OpenCV 4.10.0 Screenshots

    The images below have been resized. Click on them to view the screenshots in full size.

    OpenCV 4.10.0 Screenshot 1
  • OpenCV 4.10.0 Screenshot 2
  • OpenCV 4.10.0 Screenshot 3
  • OpenCV 4.10.0 Screenshot 4

What's new in this version:

Core Module:
- Added CV_FP16 data type for cv::Mat
- Fixed undefined behaviour in some cases related to FP16 arithmetic
- Extended HAL API for minMaxIdx
- Persistence: output reals as human-friendly expression
- Added cgroups v2 support to parallel framework
- Added in-place support for cartToPolar and polarToCart

Imgproc Module:
- Added relative displacement field option to cv::remap
- Added new findContours implementation
- Fixed allocation issue in EMD
- Fixed bug in stackBlur with large kernels
- Extended HAL for projectPoints

Calib3d Module:
- Fixed several bugs in hand-eye calibration methods
- Fixed several bugs in findHomography implementation
- Multiple improvements in findChessboardCorners implementation:
- Improved symmetric circles grid pattern detection
- Fixed possible infinite loop in USAC
- Changed approach for initial estimation for focal length in fisheye calibration
- Added solvePnP implementation for Fisheye camera model

DNN module:
- Significantly improved DNN memory consumption
- Added Net::dumpToPbtxt method to review optimized graph with Netron
- Added support for several TFLite layers: Global_Pool_2D
- Fixed bugs in several ONNX layers: Slice
- Added new ONNX layers: Mod
- Added more DNN layers with OpenVINO
- Optimized fastDepthwiseConv
- Optimized fastDepthwiseConv for RISC-V P Packed SIMD Extension v0.5.2
- Optimized Attention
- Added parallel version for scatter and scatterND
- Added broadcasting if a.rank() != b.rank() with CuDNN backend
- Tuned Winograd algorithm usage for convolutions
- Added Raft model support and tracking sample with it
- Added Vulkan backend for NaryEltwiseLayer in DNN module
- Imporved modern Yolo detectors support. Added sample and documentation
- Added CuDNN 9+ support
- Added OpenVINO 2024 support

G-API module:
- Ported G-API ONNXRT backend into V2 API
- Implement concurrent executor
- A quick value-initialization support GMat
- Fixed support of mean/scale preprocessing in OV backend for non-real types

Objdetect module:
- Fixed inaccurate Charuco board generation in some cases (.generateImage())
- Fixed Aruco detector thread safety issue
- Added QR codes Structured Append decoding mode
- Several bug fixes in QR code detector and decoder
- Fixed barcode detectAndDecode
- Ported remaning Aruco and Charuco documentation and samples to modern API

VideoIO:
- Fix race condition in InternalFFMpegRegister initialization.
- Orbbec Camera supports MacOS，Gemini2 and Gemini2L support Y16 format
- Added V4L2_PIX_FMT_SGRBG8 pixel format support to V4L2 backend
- Fix incorrect timestamps returned by VideoCapture when CAP_PROP_FORMAT == -1

Python Bindings:
- Experimental NumPy 2.0 support
- Added python bindings for Rect2f and Point3i
- Switched to numeric dtype for MatLike instead of generic
- Added path-like objects support wherever file names are expected

Android:
- Added any screen orientation support for JavaCameraView
- Removed Android AIDL from build scrips and tutorials as it's not needed since 4.9.0
- Enable file system on Emscripten
- Updated Android samples for modern Android studio. Added OpenCV from Maven support.
- Added kotlin classes to AAR
- Added ARM KleidiCV as HAL for ARM v8 and v9 processors (-DWITH_KLEIDICV=ON CMake option)

Platforms support:
- CUDA 12.4+ support
- Wayland back-end for Linux
- New LAPACK interface support on MacOS and derivatives
- Added initial version of HAL for RISC-V processors with P-extension
- Added ARM KleidiCV as HAL for ARM v8 and v9 processors (-DWITH_KLEIDICV=ON CMake option)
- Added zlib-ng as alternative to classic zlib (-DWITH_ZLIB_NG=ON CMake option)
- OneAPI 2024 support (IPP, TBB)
- Experimental Apple VisionOS support
- Experimental Windows ARM64 support

Top Downloads

More Popular Software »