Tuesday, August 31

8:00 — 9:00

Registration
9:00 — 11:00

Tutorial

Dr. David Anderson: Real-Time Processing Tutorial
11:00 — 11:20

Break

Dr. David Anderson: Real-Time Processing Tutorial
11:20 — 12:20

Tutorial
12:20 — 13:30

Lunch
13:30 — 14:30

Tutorial

Dr. Volkan Cevher and Dr. Justin Romberg: Compressive Sensing Tutorial
14:30 — 16:30

Tutorial

Dr. Volkan Cevher and Dr. Justin Romberg: Compressive Sensing Tutorial
16:30 — 16:50

Break
16:50 — 18:30

PhD Forum & Work in Progress

Resource-Aware Image Mosaicking on Networks of Small-Scale UAVs

Daniel Wischounig-Strucl and Bernhard Rinner Institute of Networked and Embedded Systems, Klagenfurt University
Abstract

In my approach nadir images taken from small-scale UAVs, are processed efficiently to estimate the depth structure of the surface to mosaick a quick overview image. Primarly only metadata and descriptors of corresponding points are transferred over the network. The image data is presented later, if sufficient communication resources are available.

vuSpot: an infrastructure for augmented information sharing using surveillance cameras

Rossana Santos and Nuno Correia CITI, FCT-UNL
Abstract

Nowadays, it is easy to find spaces with video camera networks installed. These video camera networks are mainly used for archive and surveillance but can be used for other purposes such as guiding visits, remote visiting and gaming. This work aims to propose an infrastructure to facilitate the creation of these applications. Augmented Reality provides additional information and means to create a virtual environment that meets the user's interests and encourages exploring a space. With the use of a wireless network, augmented video streams from surveillance cameras can be delivered to mobile devices. These are used to guide a visit, provide extra information or as a tool to interact with the space, other visitors and virtual objects and characters. Remote visiting in the form of telepresence can be offered by visualizing the augmented video streams, making a remote user experience as close as possible to the real visit to the space.

Automatic Hand Reference Acquisition Using a Stereo 3D Webcam

Arnaud Bernard and Benny Bing Georgia Institute of Technology
Abstract

This paper presents a framework for hand template sampling using a stereo 3D webcam and background modeling. The generated template can be used to record specific characteristics of a user’s hand or to create a hand template database.

Distributed Smart Cameras for Hard Real-Time Control

Institute of Networked and Embedded Systems, Klagenfurt University
Abstract

This paper describes the current state of the work of a system that integrates cognition, real-time communication and control. The test setup is a model crane which area of operation is observed by smart cameras which do object and motion detection. Due to the tight connection between cognition and control, information from the control system can be used to set the regions of interest of the image analysis according to the current requirements.

A Framework for Lab-based Real-time Video Analysis on Distributed Camera

Abstract

In the field of video analytics for surveillance, the trend towards the use of multi-camera and high definition video is increasing. This poses significant technical challenges in terms of video transmission and real-time processing for surveillance analytics, such as people recognition and tracking. Currently, available solutions are typically proprietary commercial systems which are costly to purchase. These proprietary systems also do not facilitate research collaboration across members of the computer vision community. We propose a framework for video analytics research based only on open-source software which is collaborative, scalable, interoperable, and distributed. This framework was successfully applied to the task of face recognition on both live video feeds and video datasets.