Upload
votu
View
217
Download
0
Embed Size (px)
Citation preview
ISCTE-IUL, Av. Forças Armadas, 1649-026 Lisboa, Portugal | Phone: + 351 210464023 | email: [email protected] | http://istar.iscte-iul.pt/
http://istar.iscte-iul.pt/
centro de investigação em ciências e tecnologias da informação e arquitetura
Propostas de Teses de Mestrado em
Computação Gráfica, Visão por Computador,
Realidade Virtual, Realidade Aumentada,
Interação Pessoa-Máquina
2016-2017
METI, MEI, MIG, MIASIC ISCTE-IUL
Prof. Miguel Sales Dias
Prof. Associado Convidado DCTI/ISCTE-IUL
Director I&D Microsoft
Prof. Sara Eloy
Prof. Auxiliar DAU/ISCTE-IUL
Diretora do Departamento de Arquitetura e Urbanismo do ISCTE-IUL
ISCTE-IUL, Av. Forças Armadas, 1649-026 Lisboa, Portugal | Phone: + 351 210464023 | email: [email protected] | http://istar.iscte-iul.pt/
Area: Virtual Reality
Topic: Understanding emotions in virtual spaces, via multi-
sensorial stimulation and biometric sensingScope• Virtual Environments are becoming increasingly popular and one niche of application that is attracting attention is
related to Virtual Environments (VEs), since they allow a museum visitor or tourist to feel immersed in sites that are
otherwise unreachable or difficult to visit, or allow a user to visit private or public spaces, before they are built. In
designing these environments, vision, hearing, touch, smell and taste are what VEs try to simulate via computing
means, so that users can report the sense of having been present somewhere else other than the place they really
were. Understanding the emotion felt by the user when navigating through the space, is also a relevant feedback to
the architect of the space. In this context, evaluating the user´s emotion and if the user felt a physical presence in a
VE is an important factor to characterize it.
Description of Work• State-of-the-art assessment
• Use biometric/ Physiological sensing available @ ISTAR-IUL (heart beat rate, electrodermal resistance, EEG -
Emotiv, eye tracking) and RGB-D sensing. Use existing VR spaces (3D building and/or urban model) with 3D
auralization and 3D sound rendering (ambisonics)
• Use several methods to measure basic emotions (anger, disgust, fear, happiness, sadness and surprise) through
Behavioral and Physiological sensing, comparison or questionnaires.
• Use CaveHollowspace middleware (CAVEH) based in OpenSceneGraph
• Perform usability evaluation in the context of students work at Architecture department @ ISCTE-IUL
• Write International/National conference paper & Thesis
Development Environment:• CAVE Hollowspace middleware based in OpenSceneGraph
• Kinect SDK, X-10 SDK, Emotiv SDK, FMOD, Processing, Biometric Sensors SDK (Plux, Ergoneers)
• Microsoft Visual Studio, C#/C++
• Blender, Matlab
Location• MLDC Porto
Ref: MSC Thesis: 3-D Sound Enhanced Presence in Virtual
Environments (António Sérgio Azevedo)
ISCTE-IUL, Av. Forças Armadas, 1649-026 Lisboa, Portugal | Phone: + 351 210464023 | email: [email protected] | http://istar.iscte-iul.pt/
http://www.rollingrains.com/2007/03/
Area: Virtual Reality
Topic: Brain-Computer Interface with Emotive• Scope
• BCI - Brain-Computer Interface
• http://www.emotiv.com/
• Emotive Provides access to dense array, high quality, raw EEG data
• 14 EEG channel locations
• 9 axis sensor (3x gyro, 3x accelerometer, 3x magnetometer)
• Description of Work • State-of-the-art assessment
• Use Emotive sensor @ ISTAR-IUL
• Conduct research using APIs and detection libraries:
• Facial Expressions
• Blink, Left wink, Right wink, Furrow (frown), Raise brow (surprise),
Smile, Clench teeth (grimace), Glance left, Glance right, Laugh, Smirk
(left side), Smirk (right side)
• Performance & Emotional Metrics:
• Instantaneous excitement, Long term excitement, Frustration,
Engagement, Meditation, Interest/ Affinity
• Mental Commands:
• Neutral, any of up to 4 pre-trained items from a list of 13 labels
• Leverage existing apps of the store
• Design the user experience
• Develop an app using Emotive SDK to detect Facial Expressions, Emotional
States or issue Mental Commands
• Perform usability evaluation
• Write International/National conference paper
• Write thesis
• Development Environment• C++, C# Microsoft Visual Studio
• Location• MLDC Porto
ISCTE-IUL, Av. Forças Armadas, 1649-026 Lisboa, Portugal | Phone: + 351 210464023 | email: [email protected] | http://istar.iscte-iul.pt/
Area: Virtual Reality
Topic: Immersive Virtual Reality with Oculus Rift• Scope
• https://www.oculus.com/rift/
• Resolution 960 x 1080 per eye
• Refresh Rate 60 Hz
• Persistence 2 ms, 3 ms, full
• 100° Field of View
• HDMI 1.4b
• USB 2.0
• Gyroscope, Accelerometer, Magnetometer
• 1000 Hz
• Positional Tracking with Near Infrared CMOS Sensor (60 Hz)
• Description of Work • State-of-the-art assessment
• Use Oculus Rift @ ISTAR-IUL
• Test existing VR apps available in the Oculus Rift store, using 3D desktop @
ISTAR-IUL VR Lab
• Integrate Oculus Rift in in-house VR system
• Leverage existing apps of the store
• Design the user experience
• Develop an app using Oculus Rift
• Perform usability evaluation
• Write International/National conference paper
• Write thesis
• Development Environment• CaveHollowspace (CAVEH) @ ISTAR-IUL with FMOD 3D sound, based in
OpenSceneGraph
• Unity 3D
• C++, C# Microsoft Visual Studio
Location• MLDC Porto
ISCTE-IUL, Av. Forças Armadas, 1649-026 Lisboa, Portugal | Phone: + 351 210464023 | email: [email protected] | http://istar.iscte-iul.pt/
• Scope• VR usually defines a set of computer technologies that can be used to create
simulated environments, with several levels of realism. CAVE (Cave Automatic
Virtual Environment) is a multi-projection immersive virtual reality environment
where projectors are directed to several walls of a room-sized cube. BIM (Building
information modeling) is a process involving the generation and management of
digital representations of physical and functional characteristics of a facility, using
standards for the definition and exchange of product data.
• Description of Work• State-of-the-art assessment
• Specify a system allowing to exchange BIM data with the existing CaveH
environment. (Architecture department @ ISCTE-IUL provides the 3D building
model)
• Extend in-house VIARModes system @ ISTAR-IUL
• Recognize gestures via RGB-D sensing (Kinect One)
• Use European Portuguese speech API
• Develop a real-time prototype
• Perform usability evaluation in the context of students work at Architecture
department @ ISCTE-IUL.
• Write International/National conference paper
• Write thesis
• Development Environment• Revit or other BIM system
• CAVE Hollowspace middleware based in OpenScenegraph
• Blender
• Microsoft Visual Studio C#/C++, Kinect SDK
• Location• MLDC Porto
Area: Virtual Reality
Topic: Use of BIM in an immersive virtual reality environment
ISCTE-IUL, Av. Forças Armadas, 1649-026 Lisboa, Portugal | Phone: + 351 210464023 | email: [email protected] | http://istar.iscte-iul.pt/
• Scope• VR usually defines a set of computer technologies that can be used to create
simulated environments, with several levels of realism. CAVE (Cave Automatic
Virtual Environment) is a multi-projection immersive virtual reality environment
where projectors are directed to several walls of a room-sized cube
• Description of Work• State-of-the-art assessment
• Specify a system allowing the edition (moving, stretching, scaling, erasing, etc)
of a 3D model of building, within immersive VR and using natural gestures and
speech interaction (Architecture department @ ISCTE-IUL provides the 3D
building model)
• Recognize gestures via RGB-D sensing (Kinect One)
• Use European Portuguese speech API
• Use CaveHollowspace (CaveH) @ ISTAR-IUL
• Develop a real-time prototype
• Perform usability evaluation in the context of students work at Architecture
department @ ISCTE-IUL.
• Write International/National conference paper
• Write thesis
• Development Environment• CAVE Hollowspace middleware based in OpenScenegraph
• Blender
• Microsoft Visual Studio
• C++, Python, Kinect SDK
• Location• MLDC Porto
Area: Virtual Reality
Topic: 3D model edition in immersive virtual reality environment
ISCTE-IUL, Av. Forças Armadas, 1649-026 Lisboa, Portugal | Phone: + 351 210464023 | email: [email protected] | http://istar.iscte-iul.pt/
Area: Human-Robot Interaction
Topic: Communication between elderly and
Robot @ Home
Description of Work • State-of-the-art assessment
• Specify an interactive experience between a robot and an elderly
person to assist @ home
• Develop a NAO app using visual programming with Choreographe
• Use NAO Robot @ ISTAR-IUL
• Leverage existing NAO apps https://www.aldebaran.com/en
• Interaction:
• Loudspeakers
• Microphones
• Video camera
• Infra-Red
• LEDs
• Sensors:
• Force Sensitive Resistors
• Inertial unit
• Sonars
• Joint position sensors
• Contact and tactile sensors
• Design the user experience
• Develop a NAOapplication
• Perform usability evaluation
• Write International/National conference paper
• Write thesis
Development Environment
https://www.youtube.com/watch?v=nNbj2G3GmAo
SDK C++ e Pyhton
Choreographe
App store
Location: ISCTE-IUL, Lisbon
ISCTE-IUL, Av. Forças Armadas, 1649-026 Lisboa, Portugal | Phone: + 351 210464023 | email: [email protected] | http://istar.iscte-iul.pt/
Area: Image Processing & Computer Vision
Topic: Large RGB-D Full Object Dataset• Scope
RGB-D object datasets are an essential requirement for the computer
vision community, for tasks such as object categorization, instance
recognition and pose estimation. The availability of 3D object models
registered and grouped by category is a novel interesting R&D hypothesis
for the research community.
• Description of Work State-of-art (SOA) assessment
Collect household 3D objects using available ISCTE-IUL data collection
system, with Kinect One Sensor for Windows: per object , collect depth images,
camera pose and compute global point cloud
Develop a global polygon mesh reconstruction algorithm using SOA techniques
Develop an Azure-based database to host the collected objects
Evaluate state-of-the-art of object classifiers and develop an object recognition
algorithm
Leverage current experience of the team with Eurographics Workshop on 3D
Object Retrieval (2015) - Shape REtrieval Contest (SHREC’15) Track: Retrieval
of Objects Captured with Kinect One Camera
Organize an international completion (2016) in the framework of Shape
REtrieval Contest (SHREC) - http://aimatshape.net/event/SHREC
Write International/national conference paper & thesis
• Development Environment C++, Microsoft Visual Studio
Point Cloud Library
Artoolkit
http://www.cs.washington.edu/rgbd-dataset/
• Location• ISCTE-IUL, ISTAR-IUL, Lisbon
ISCTE-IUL, Av. Forças Armadas, 1649-026 Lisboa, Portugal | Phone: + 351 210464023 | email: [email protected] | http://istar.iscte-iul.pt/
Area: Mobility Apps
Topic: Augmented Reality App for Windows 10• Scope
• AR is the provision of intuitive information presentation, which enhances
the perceiver’s situational awareness and cognitive perception of the real
world. Enhancement is achieved by placing virtual objects or information
cues into the real world. AR (Azuma 97): (a) combines real and virtual
environments; (b) is interactive in real-time; (c) is registered in 3D
• Description of Work• State-of-the art assessment
• Specify, develop and test an AR prototype for mobility (Windows 10)
based on FIRST (Fast Invariant to Rotation and Scaling feature
Transform) tracking system available @ ISTAR-IUL
• Improve existing in-house developed AR system Arch
• Conduct a benchmark evaluation between the system developed and
other state-of-the-art techniques
• Conduct an usability study of the developed prototype in Windows 10 with
a mobility AR application for: (i) a museum which acts like a complement
to the exhibition showing additional information, or ii) a system for city
map travel (Architect department @ ISCTE-IUL provides the 3D building
models and other elements)
• Write thesis and national/international conference paper
• Development Environment• C# and Win10 SDK
• Universal App framework
• Microsoft Visual Studio
• OpenCV
• Location
• ISCTE-IUL
• ISTAR-IUL, Lisbon