Displaying 1 to 4 from 4 results

sod - An Embedded Computer Vision & Machine Learning Library (CPU Optimized & IoT Capable)

  •    C

SOD is an embedded, modern cross-platform computer vision and machine learning software library that expose a set of APIs for deep-learning, advanced media analysis & processing including real-time, multi-class object detection and model training on embedded systems with limited computational resource and IoT devices. SOD was built to provide a common infrastructure for computer vision applications and to accelerate the use of machine perception in open source as well commercial products.

pico - A minimalistic framework for fast object detection (with a pre-trained face detector)

  •    C

Those of you who would like to quickly see what this repository is all about, go to the folder rnt/sample. There you will find a sample program which will detect faces in a video stream supplied from the default webcam attached to the computer. Also, you can check out a demo video at http://www.youtube.com/watch?v=1lXfm-PZz0Q. In general, detection can be described as a task of finding the positions and scales of all objects in an image that belong to a given appearance class. For example, these objects could be cars, pedestrians or human faces. Automatic object detection has a broad range of applications. Some include biometrics, driver assistance, visual surveillance and smart human-machine interfaces. These applications create a strong motivation for the development of fast and accurate object detection methods.

RPIMotionDetection - Blob detection library with example applications

  •    C

If you compile on the RPi: Please note that the compiling of the release version (-O3) requires a big amount of memory. If the compiling fails reduce the memory for the gpu at 64MB (config.txt) or enable swapping. If you compile on other systems: Add the cmake flag -DWITH_RPI=0 to disable the parts which requires RPi dependecies.

MutScan - Detect and visualize target mutations by scanning FastQ files directly

  •    C

-t argument specify how many worker threads will be launched. The default thread number is 4. Suggest to use a number less than the CPU cores of your system.