Sod/README.md

# SOD - An Embedded Computer Vision & Machine Learning Library - [sod.pixlab.io](https://sod.pixlab.io)
![Output](https://i.imgur.com/YIbb8wr.jpg)
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.

Designed for computational efficiency and with a strong focus on real-time applications. SOD includes a comprehensive set of both classic and state-of-the-art deep-neural networks with their <a href="https://pixlab.io/downloads">pre-trained models</a>. Built with SOD:
* <a href="https://sod.pixlab.io/intro.html#cnn">Convolutional Neural Networks (CNN)</a> for multi-class (20 and 80) object detection & classification.
* <a href="https://sod.pixlab.io/api.html#cnn">Recurrent Neural Networks (RNN)</a> for text generation (i.e. Shakespeare, 4chan, Kant, Python code, etc.).
* <a href="https://sod.pixlab.io/samples.html">Decision trees</a> for single class, real-time object detection.
* A brand new architecture written specifically for SOD named <a href="https://sod.pixlab.io/intro.html#realnets">RealNets</a>.

![Multi-class object detection](https://i.imgur.com/Mq98uTv.png)

Cross platform, dependency free, amalgamated (single C file) and heavily optimized. Real world use cases includes:
* Detect & recognize objects (faces included) at Real-time.
* License plate detection.
* Intrusion detection.
* Mimic Snapchat filters.
* Classify human actions.
* Object identification.
* Eye & Pupil tracking.
* Facial & Body shape extraction.
* Image/Frame segmentation.

#### Notable SOD features
* Built for real world and real-time applications.
* State-of-the-art, CPU optimized deep-neural networks including the brand new, exclusive <a href="https://sod.pixlab.io/intro.html#realnets">RealNets architecture</a>.
* Patent-free, advanced computer vision <a href="https://sod.pixlab.io/samples.html">algorithms</a>.
* Support major <a href="https://sod.pixlab.io/api.html#imgproc">image format</a>.
* Simple, clean and easy to use <a href="https://sod.pixlab.io/api.html">API</a>.
* Brings deep learning on limited computational resource, embedded systems and IoT devices.
* Easy interpolatable with <a href="https://sod.pixlab.io/api.html#cvinter">OpenCV</a> or any other proprietary API.
* <a href="https://pixlab.io/downloads">Pre-trained models</a> available for most architectures.</li>
* CPU capable, <a href="https://sod.pixlab.io/c_api/sod_realnet_train_start.html">RealNets model training</a>.
* Production ready, cross-platform, high quality source code.
* SOD is dependency free, written in C, compile and run unmodified on virtually any platform &amp; architecture with a decent C compiler.
* <a href="https://pixlab.io/downloads">Amalgamated</a> - All SOD source files are combined into a single C file (*sod.c*) for easy deployment.
* Open-source, actively developed & maintained product.
* Developer friendly <a href="https://sod.pixlab.io/support.html">support channels.</a>

![face detection using RealNets](https://i.imgur.com/ZLno8Lz.jpg)