The article describes the making of an art piece combining stone sculpture and virtual water. The motivation for this art piece was to enrich the usual static format of a stone sculpture with a dynamic dimension. The dynamic dimension is attained with virtual water droplets running over the stone surface which behave as real water droplets. The 3-D surface of a specially carved stone sculpture is during an exhibition continuously captured by the Kinect sensor. Each water drop out of many thousands, which are introduced into the installation as evenly distributed rain drops, falling over the sculpture, are simulated individually to run over the stone surface following the largest slope. These simulated water drops are projected with a video projector as light points on the surface of the sculpture. An observer can enjoy simultaneously the haptic experience of touching the stone and observing a digitally generated but physically grounded animation. https://youtu.be/nqituTE-_SI?list=PLHGd9WmePQ0g7H9fPM3iZ6J_Nep2Bm2lH
F.17 Transfer of existing technologies, know-how, methods and procedures into practice
COBISS.SI-ID: 1537281219Radio emission from the series A view into science. Guest of today's emission was prof. dr. Franc Solina. We were discussing possibilities how computers could see as human eyes can observe. The discussion was centered around the challenges of this important aspect of artificial intelligence. The research group led by prof. Solina tries to verify theoretical results on different existing platforms such as mobile robots, active sensory systems and smart mobile devices. The series is prepared by Goran Tenze.
F.30 Professional assessment of the situation
COBISS.SI-ID: 1536769219After a long period of learning photogrammetric methodologies for documentation of underwater archaeological sites, which we started in eastern Adriatic 15 years ago on the island of Silba, we finally got an opportunity to replace old site documenting methods based on measuring with contemporary possibilities offered by rapidly progressive development of 3D documentation. Despite our desire to carry out simultaneously also the previous standard documentation procedure using tape measurement to demonstrate the many advantages of using 3D documentation, due to the lack of time and adverse weather conditions it had to be abandoned. On this early Roman barge site in river Ljubljanica we demonstrate and prove that the new methodology of documenting underwater sites extremely facilitates archeological fieldwork. As it was expected, it turned out that contemporary approaches are much cheaper due to significantly shorter underwater data acquisition, more accurate documentation and due to shorter diving times required for data acquisition also greatly increased safety.
F.17 Transfer of existing technologies, know-how, methods and procedures into practice
COBISS.SI-ID: 1537311683In 2016 program group members co-organized the twenty first edition of Computer Vision Winter Workshop, CVWW2016. CVWW is a yearly conference organized by research institutions from Slovenia, Austria and Czech Republic. In 2016 the conference was organized by the Slovenian pattern recognition society in Rimske Toplice; 40 researchers from different countries have presented 22 computer vision and robotics papers.
B.01 Organiser of a scientific meeting
COBISS.SI-ID: 283229440Program group members are among the lead organizers of short-term visual object tracking challenges, VOT. The VOT project committee is international and includes partners from Slovenia, Great Britain, Austria, Czech Republic, Australia and Sweden. The project leader is the lead methodologist, analyst and scientific coordinator in the committee. Apart from the challenges he coorganized workshops at main conferences of computer vision, VOT2013 at ICCV2013, VOT2014 at ECCV2014, VOT2015 at ICCV2015, while the latest challenge VOT2016 was organized in conjunction with ECCV2016. The challenges as well as workshops have received a highly positive response from the community. A large number of authors from industry as well as academia have submitted their results to the most recent challenge. The results have been published in over 140 coauthors paper. The methodology developed within the VOT is becoming a standard in tracker analysis in the field of computer vision.
B.04 Guest lecture
COBISS.SI-ID: 1537235139