home

This is the official Wiki page of the EAVISE research group.

EAVISE (**E**mbedded **A**rtificially intelligent **VIS**ion **E**ngineering) is a multidisciplinary research group of KU Leuven based at Campus De Nayer in Sint-Katelijne-Waver focusing on applications of advanced computer vision and artificial intelligence. The research goal of EAVISE is applying state-of-the-art computer vision techniques as a solution for industry-specific vision problems. In order to meet stringent execution speed, energy footprint, cost and price requirements, the developed algorithms are implemented and optimized on embedded systems, such as GPUs, DSPs and FPGAs.

Application areas include industrial automation, product inspection, traffic monitoring, e-health, agriculture, eye-tracking research, microscopic imaging, surveillance and land surveying.

The vision part of EAVISE is integrated with [|VISICS]. The Artificial Intelligence part of EAVISE is integrated with [|DTAI].

This Wikispace is meant to document different software and hardware problems we've overcome. In order to help future researchers who are having the same problems, all solutions are documented.

=__Frameworks__=
 * //ACF pedestrian detector - Piotr Dollar//**
 * Installation of the toolbox and code base - ACF Piotr Dollar
 * Training your own model - ACF training
 * Creating PR curves of your model - ACF PR Curves

=__OpenCV related HOW-TO's__= Linux based solutions:
 * OpenCV installation on Linux systems for OpenCV 2.0, OpenCV 2.3.1
 * OpenCV installation with CUDA support for OpenCV 2.4.3 (Ubuntu 12.04.3 LTS)
 * OpenCV installation with CUDA 6.0 (and 6.5) on Ubuntu 14.04
 * Example of a make file
 * Installation of mexopencv
 * Installation of a USB camera, a FireWire camera or a Composite USB interface

Windows based solutions:
 * Configuring OpenCV on W7 x64 with VS2010 for [[file:OpenCV - Configure New Project - Windows 7 x64.pdf|OpenCV 2.3.1]]
 * Building manually OpenCV 2.4.5 W7 x64 CUDA 5.0

=__Embedded platform related HOW-TO's__= BeagleBoard solutions can be found here. PandaBoard solutions can be found here. DM 816x DSP Board solutions can be found here.

Installation of packages
 * Guidelines for installing Xilinx Design Suite (Ubuntu)

=__Hardware related HOW-TO's__= Hardware guidelines and solutions can be found here.
 * Manta GigE camera
 * Guppy met Coriander
 * Combining IDS UI-622 camera with OpenCV 2.4.9

=__Software (OS + packages) related HOW-TO's__=
 * MatLab
 * How to get the VeryFast-framework (object detection) to run (Ubuntu 12.04)
 * Install Ubuntu 12.04.3 + CUDA 5.5 on laptop with NVIDIA OPTIMUS technology
 * Ffmpeg tips & tricks

=__Simple & small guides and tricks__=
 * Filmpjes maken aan de hand van afbeeldingen
 * Afbeeldingsreeksen bewerken
 * Frames uit film halen
 * Youtube heeft blauwe kleur in Ubuntu 12.04
 * Debuggen en profilen van C en c++ applicaties
 * Debuggen van mex-code met gdb
 * Papers downloaden
 * Using git version control on NAS
 * De netwerkprinters installeren en gebruiken

=__Administration__=
 * Verplaatsing aanvragen in KU Loket
 * Onkostennotas verplaatsing (eigen auto + openbaar vervoer) en andere kosten + sporadisch gebruik trein woon-werk
 * Posters laten afdrukken
 * Reis boeken via Omnia
 * Hardware bestellen

=__Current project wiki's and information websites__=
 * Sivol
 * SOSOpenCL
 * TOBCAT
 * 3D4SURE