UrbanSensing

Started in 2014 we deployed a large scale 3D visual sensing infrastructure in Lisbon’s Airport. Currently with more than 30 depth cameras (3D sensors), became a key technology for queue management in   border control and security checkpoints. Fully developed at SIPg@LarSys was transferred to Thales Group that operates and maintains the infrastructure. It provides high-level analytics, extracting queue operating patterns from visual motion analysis and 3D tracking data. Hinging on sparse data representations and efficient convex optimization algorithms it subsumes part of our f

rontier research: i) low-rank and sparse models with missing data,  ii) large scale optimization algorithms and iii) 2D and 3D computer vision algorithms for tracking and

recognition. This research was published in top journals IEEE-PAMI, several IEEE-CVPR (highest ranked publication in Pattern Recognition) and summarized in IEEE-ITS (archive.org). It contributed strongly to the establishment of Thales’ innovation hub in Lisbon who currently sponsors several of our research projects for “smart-cities” (smartbyke, walkbot, smartcitysense)

LARSyS 2018 Annual Meeting

The 2018 LARSyS Annual Meeting was a success!

For two full days researchers from ISR, IN+, M-ITI and Maretec gathered at Pavilhão do Conhecimento to exchange knowledge and debate future challenges and opportunities. Take a look at some of what went on during all these vibrant activities by checking the LARSyS Flickr here or explore the recently live website of the Associated Laboratory, for more information.

See details in Link

IBM Scientific Awards

Pinar Oguz Ekim is the second woman to receive the IBM Scientific Award in 23 years. The award ceremony takes place this Tuesday in Lisbon.

The award, worth €15,000, “aims to distinguish the contribution of research work to the development of computer science and information technology,” IBM said in a statement.

The work of Pinar Ekim, from the Institute of Systems and Robotics (Instituto Superior Técnico), is entitled “Robust location algorithms in sensor networks with target tracking applications”, is based on the PhD thesis of the researcher and addresses” the problem for the determination of geographical positions of one or more agents (eg people, vehicles or animals) from measures of mutual distance and distances to reference points “.

According to the company, “this work introduces a new approach in how, in places where it is not possible to access the GPS system (either by equipment limitations or by the absence of GPS signals indoors), alternative technologies can be used to measure distances using acoustic signals, radio signals, or both”.

See details in Link