libccv

Index

1. Introduction
2. Limitations
3. Dependencies
4. Licensing

Introduction

libccv is a minimalist open-source computer vision library for embedded devices developed by Liu Liu. It is meant to be easy to deploy and has a simple and well-organized code structure.

Its main characteristics are:

1. It is portable and embeddable.
2. It has a clean interface with cached image pre-processing.
3. It includes several modern computer vision algorithms, for example:
   • Image classifier
   • Frontal face detector, object detectors for pedestrians and cars
   • Text detection algorithm
   • Object tracking algorithm
   • Feature point extraction algorithm

Detailed information of all the functionalities of the library can be found in http://libccv.org/

Limitations

The library is not currently optimized for Myriad 2, using only the LeonOS processor.

There are some known limitations of the port (which work in the original libccv):

1. JPEG is not supported
2. Reading/writing sqlite3 files is not supported. This feature could be used in read/write methods of:
   • ccv_scd
   • ccv_convnet

Dependencies

1. zlib
2. cblas
3. libpng

Licensing

Libccv source code is distributed under BSD 3-clause License.

Files in directories ./samples are licensed under the Creative Commons Attribution 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Copyright (c) 2010, Liu Liu
All rights reserved.