• No results found

PROJECT: “Geometrically Enriched Computer Vision Algorithms For 3D/4D Reconstructions”

N/A
N/A
Protected

Academic year: 2021

Share "PROJECT: “Geometrically Enriched Computer Vision Algorithms For 3D/4D Reconstructions” "

Copied!
1
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

ITN-DCH TIMELINE

PROJECT: “Geometrically Enriched Computer Vision Algorithms For 3D/4D Reconstructions”

Supervisor: Prof. Luc Van Gool

Catholic University of Leuven (KUL)

Gina Stavropoulou, ESR5

EDUCATIONAL BACKGROUND

SECONDMENTS:

RESEARCH

Case Study: Byzantine Church of Asinou

WORKING PACKAGES

KUL is responsible for WP3

Deliverables: 3.1, 3.2, 3.4

Home Country Host Country Host Organization

1 2 3 4 5 6 7 8 9 10 11 12

Oct. 2013 Jan. 2014

Jan. 2015

Euromed Conference Cyprus

Summer School Stuttgart

CNRS secondment

KAAK Secondment Carnuntum

Workshop Recruitment

Oct. 2014

SEAHA Conference Summer School

Rhodes

Researcher’s Night

13 14 15 16 17 18 19 20 21 22 23 24

Now

MSc in GIS &

Archaeology Dipl. Eng. In Rural and

Surveying Engineering

Data: Images of the Frescoes of the Asinou Church’s Interior.

Mainly religious figures with similar characteristics (halos, hand gestures, clothing)

Explore different feature sets for the fast detection of inter-fresco c o r r e s p o n d e n c e s , a n d , e v e n t u a l l y d e t e c t s i m i l a r images/figures.

2015:

•  CNRS, Marseille

(May 4th-May 29th)

•  KAAK, Bonn

(June 16th-July 3rd)

2016 :

•  Fraunhofer, Darmstadt

•  ArcTron, Altenthann

(Dates to be determined)

FUTURE EVENTS:

SEAHA 2015

Researcher’s night Fill out missing parts, due

to cracks and other d e t e r i o r a t i o n s , u s i n g content-aware, texture synthesis or inpainting techniques, taking into account information from similar images.

The central goal of the research is to explore how computer vision methods of 2D/3D shape retrieval, reconstruction and completion can be enriched and implemented to assist cultural heritage applications.

At the moment the work is focused on feature

extraction algorithms and more specifically on their

implementation on 2D data collected from the

byzantine church of Asinou in Cyprus, one of the

project’s case studies. Additionally, 3D feature

extraction algorithms are being explored with

implementation on 3D data of small archeological

objects collected with the Minidome of the Visics

group in KUL.

Referenties

GERELATEERDE DOCUMENTEN

It is still challenging to recognize faces reliably in videos from mobile camera, although mature automatic face recognition technology for still images has been avail- able for

The questions respectively concern ”How [...] a wearable computer vision system [would] need to be designed to detect important features to aid somebody with a visual

The most important recommendation is that Faster R-CNN needs to be trained with images from real broken billboards in order to be effective.. Additionally, training with

J.. Such systems are mainly used in coffee and milk foam treatment. PCV’s systems are analysed intensively to ensure their quality. Currently, their milk foam analysing

A discussion session held at the AARG annual conference during September 2016 in Plzeň on the topic of ‘automation’ addressed this issue through three papers: Dimitrij Mlekuz

With two series of experiments designed, a model-based recognition algorithm and an image-based recognition algorithm are applied to find the difference in object

Although effects of N deposition on the loss of bare sand in inland dunes were partly masked by differences in geomorphology, it has been shown that -on a local scale-

However, it has to be noticed that once PDAs trigger plant responses, roots can release signal molecules such as salicylic acid (SA) ( Cheol Song et al., 2016 ) to induce plant