[visionlist] Post-doc: Visual landmark detection and matching for image-based localization (Paris area, France)Posted: November 30, 2012
Title : Visual landmark detection and matching for
image-based localizationLink to the offer:
This Postdoctoral position will take
place within the scope of the large TerraMobilita project
(a follower of
the TerraNumerica project) of the French CapDigital
cluster (Business Cluster For Digital Content and
Services of the Paris Region) which aims at building
production lines to generate automatically and semi-
automatically 3D models of roadways and sidewalks from
ground-based mobile mapping systems (MMS)
equipped with optical imaging together with laser devices.
In that scope, one technological lock is to
obtain a high quality localization of the acquired data,
enabling for instance robust map updating by
change detection when passing several times by the same
path at different dates. This high quality can
be achieved using Inertial Navigation Systems (hybridating
GPS, INS and odometer measurements) but
at a very high cost. It impedes the exploiting or
development of low-cost digitizing systems that are able
acquire renewed data with very short updating cycles to
ensure the temporal integrity of the data basis.
An alternative strategy is to use the optical sensors of a
light device (a light mobile platform, a
Smartphone or a simple camera) as positioning sub-system
and to register the acquired data with respect
to a reference data set that would be acquired by an
infrastructure vehicle integrating a high quality
localization device. Here, the infrastructure acquisitions
will be performed with a mobile mapping system
called STEREOPOLIS which is developed in the MATIS
laboratory. The images acquired by the low-cost
system can then be registered by first matching visual
relevant features between both data sets, and
second integrating these matches into a bundle adjustment
process. The MATIS has already some
experience and tools on bundle adjustment for robust
registration and reconstruction of 3D visual
landmarks (e.g. road markings and road signs) from
multiple view imagery with sub-decimetric absolute
In this context, the post-doc will focus
on the study of matching algorithms for very precise
an image as well as of an image sequence. The objective is
to increase the accuracy already reached in
the laboratory and in literature. In particular, we will
first focus on the extraction of very precise visual
features. Visual landmarks such as road markings and signs
have proved their robustness, especially
facing classical interest points. Other visual landmarks
or more general visual features (points, lines,
patches etc.) will be studied according to the precision
of their localization and their repeatability.
Then the main challenge will concern the
geometric precision of the matching, especially facing the
volume and redundancy of the manipulated data (input and
reference) and their differences due to
different periods of acquisition. Pure 2D-to-2D strategies
may be considered, as well as 2D-to-3D ones
involving the 3D reconstruction associated with the
reference dataset available at MATIS. Depending on
the source of the images, an initial set of solutions
(e.g. given by GPS) would be available or not. In that
case, the scalability of the proposal will be mandatory to
deal with large segments of data.
The MATIS laboratory of the IGN, which
is the French national mapping agency, is one of the
laboratories in photogrammetric computer vision, image
analysis and remote sensing applied to
geospatial imagery and ground based imagery (e.g.,
provided by mobile mapping systems). It is
composed of 30 researchers, including 17 permanent
researchers. The MATIS laboratory has
been involved in 3D data collection for 3D city modelling
for twenty years, and
makes use of several distinct methods that have been
developed during this period. For more information
about the MATIS please visit our website:
The candidate should have a PhD degree
in photogrammetry or computer vision, with knowledge and
interest in pattern recognition.
– Good knowledge of programming language
(C++) is mandatory.
– Prior knowledge and experience in the fields of pose
estimation and/or scalability will be a plus.
– Good spoken and written English. Knowledge of French
would be useful.
Location: MATIS laboratory of the IGN,
Saint-Mandé, Paris, France.
Salary: around 2200 € per month (net income), according to
experience. The position is a salaried
employment with the right to social benefits and paid
Duration: 18 months, to start as soon as possible.
Send by email in a single pdf file to
the two contacts: a cover letter describing how your
experience is relevant to the position, recommendation
letters or names of referees and a resume
(including a summary of the thesis).
Deadline: Position is open until filled.
– Bahman Soheilian Phone: 00 33 1 43 98
84 29 / E-mail: email@example.com
– Valerie Gouet-Brunet Phone: 00 33 1 43 98 00 00 /
A postdoctoral position at Monash University,
ARC (Australian Research Council) funded research
project on â€œThe neuronal basis of consciousness: how brain rhythms control the doors
of perceptionâ€ (PIs: Tsuchiya, Maller, Foster, Takaura), which will run for 3
years from Jan 2013. The initial appointment will be for one year (AUD $72,756/per year), with renewal based upon satisfactory performance and evaluation.
We are looking for a
postdoctoral fellow whose main research interest is the neuronal basis of
consciousness. In particular, we look for applicants who have experience and
publications in EEG and TMS experiments as well as experience (and possibly
publications) in advanced signal processing analyses of the multi-dimensional data
in EEG and ECoG, such as Granger Causality and decoding/machine learning. Experience
and publication in the advanced analysis of eye movements (micro-saccades, scan
path analysis, etc) would be also welcomed. It is essential to have programming
skills in Matlab for task design, data analysis, and statistics.Â
Excellent written and oral English communication skills is required.
We will be mainly
working at the Clayton Campus of Monash University, one of the eight top
universities in Australia (G8). Depending on the progress of the project, the
initial contract will be extended to involve simultaneous EEG & fMRI
experiments and EEG/TMS & fMRI experiments at Monash Brain Imaging
facility, which hosts a research-only MRI scanner, EEG, TMS and eye tracking.Â http://mbi.monash.edu.au/
are encouraged to send a CV, statement of research accomplishments and
interests, PDF copies of representative publications, and contact information
for 2 reference letters to:
Application review will begin immediately and
continue until the position is filled.
I would appreciate for the following vacancy to be posted on Imageworld.
SeeByte provides clients in the Military and Oil and Gas sectors with the most advanced software to enhance the capabilities of their underwater sensors, vehicles and systems.
SeeByte are looking to recruit a Graduate Test Engineer to join their Edinburgh based Engineering group.
Graduate Test Engineer – Edinburgh
Salary – Dependant on skills/experience
Company – SeeByte Ltd (www.seebyte.com)
Job Type – Permanent
As a Graduate Test Engineer, your main responsibilities will involve:
• Ability to follow the appropriate methodology in order to build, execute and analyse functional and non-functional tests.
• Implementing test tools, environments and test data in a timely and cost effective manner.
• Executing and analysing test cases and scenarios.
Academic Qualifications / Experience
• Good degree (1st /2.1 Hons) Electrical/Computer Science/Software Engineering with strong software engineering emphasis.
• Practical experience in working on Linux and Windows systems.
Essential Technical Skills
Demonstrable experience in:
• General software engineering skills
• Unit testing and functional testing
• Experience in at least one of the following areas:
◦ Scripting and task automation
◦ Code Profiling and performance analysis
◦ System integration and safety assurance
• Excellent organisational skills.
• Good inter-personal/communications skills.
• Confident working in a small company environment.
• Good reporting skills.
• Dependable when working under tight deadlines.
How to Apply
Send an e-mail to firstname.lastname@example.org in person (no recruitment agencies) with an up to date CV, including current salary details and notice period.
Because of the nature of the work associated with this post, it is subject to special nationality rules and is open only to British, US or European Citizens. All selected candidates will be subject to background checking.
HR Account Manager
14 India Street
Office: 0131 225 7458
New direct dial: 0131 243 1377
Office Mob: 07767819568 / 07767819567
Registered in Scotland No SC287870. Registered Office 5th Floor, Quartermile Two, 2 Lister Square, Edinburgh, EH3 9GL. VAT Reg. No. 873 5229 06
This email and any files Click here to read our full e-mail disclaimer
[Imageworld] Call for Participation: 2nd Competition on Counter Measures to 2D Facial Spoofing AttacksPosted: November 30, 2012
2nd Competition on Counter Measures to 2D Facial Spoofing Attacks –
CALL FOR PARTICIPATION
The Tabula Rasa project and the Idiap Research Institute are
organizing the 2nd Competition on Counter Measures to 2D Facial
Spoofing Attacks. It will be organized as a part of the 6th
International Conference on Biometrics, ICB-2013, to be held in June
4-7, 2013 in Madrid, Spain.
Please find the Call for Participation here: http://www.idiap.ch/~ichingo/resources/ICB2013-antispoofing-competition.pdfImportant dates:
Registration due: January 14, 2013
Availability of test set: March 1, 2013
Submission of results and method description: March 15, 2013
Publication of the results at ICB-2013: April 8, 2013
For information about the competition, please visit: http://www.tabularasa-euproject.org/evaluations/icb-2013-face-anti-spoofing
The Robotics Group at the University of Zaragoza, Spain is looking for two highly motivated PhD students to join an industry-driven research project in collaboration with a recently created start-up. The post is available from January 2013. The successful candidate must have a degree(s) in Engineering, Physics, Maths, Computer Science or a related field, with a solid background in machine learning and/or computer vision. The PhD students will work with Dr. Luis Montesano and Dr. Javier Civera.
Applicants should submit a cover letter addressing the selection criteria below, curriculum vitae (including publications), and contact information of at least one referee. A copy of the academic transcripts (i.e., your grades in your bachelor and/or master degree(s)) should be included in the application. Applications should be sent to Â email@example.comÂ in a single pdf document; indicating which one of the two positions is preferred. Applications can be sent immediately and will be evaluated until the position is taken.
â€¢ Â Â Â Excellent C/C++ and Matlab programming skills.
â€¢ Â Â Â Excellent academic background. A Master degree in related fields will be highly valued.Â
â€¢ Â Â Â Experience in machine learning methods.
â€¢ Â Â Â Experience in computer vision (3D vision and/or recognition).
The project aims to create portable technology to evaluate the behavior of people on everyday situations, such as shopping.
Profile 1: 3D scene models from wearable cameras. The aim is camera tracking and the creation of scene models from an image sequence obtained with a wearable camera. The specific challenges that will be addressed are: blurred images from rapid motion, scenes with low texture and the estimation of high level semantics (e.g., inserting objects or humans in the scene). Â This PhD thesis will be mainly advised by Dr. Javier Civera.
Profile 2: Automatic creation of user profiles from behavioral data. The research will focus on the development of algorithms to extract user profiles from real data acquired during everyday activities. Data will involve many different measures including trajectories, attention and information provided by bio-sensors. This PhD thesis will be mainly advised by Dr. Luis Montesano.
[Imageworld] Call for Participation: The ICB-2013 Competition on Face Recognition in Mobile Environment using the MOBIO DatabasePosted: November 30, 2012
Call for Participation
In the context of BEAT project, the Biometric group at the Idiap
Research Institute is organizing the second competition on face
recognition for the 2013 International Conference on Biometrics
(ICB-2013) to be held in Madrid, Spain on June 4-7, 2013. Researchers in
biometrics are highly invited to participate in this competition. This
will help them to evaluate the progress made in the last couple of years.
The competition will be carried out on the MOBIO database. MOBIO is a
challenging bimodal (face/speaker) database recorded from 152 people
using mobile phones and laptop computers. Hence, algorithms that perform
well on this database are highly probable to be suitable for other
real-world applications that do not require a predefined image recording
More information can be found in the call for participation:
and in the web-page of the competition:
For further information, please contact:
Call for papers for a special issue on “Visual Control of Mobile
Robots” from the Journal of Electrical and Computer Engineering
Visual control of mobile robots (ground, aerial, underwater, and
humanoid robots) refers to the use of visual signal analysis and
processing techniques to achieve autonomous control and navigation of a
mobile robot. This is a multidisciplinary area that integrates
techniques from various disciplines such as signal processing,
image/video processing, computer vision, and control to increase the
autonomy of mobile robots. The importance of this area lies in the many
potential applications such as industrial automation, material
transportation, inspection, surveillance, search and rescue, assistance
to humans, and critical infrastructure protection, to mention a few.
This special issue intends to provide researchers, academics, and
industrialists, with an update of the most recent research results,
developments, and applications in the area of visual control of mobile
robots. Potential topics include, but are not limited to:
Visual sensors and systems for control of mobile robots
Integration of visual information in the control loop
Visual perception techniques for control of mobile robots
Autonomous navigation and visual servoing techniques for mobile robots
Tracking, scene analysis, path planning, and obstacle avoidance for
Distributed visual control of mobile robots
Multirobot communication and coordination
Robustness and stability issues of visual control of mobile robots
Implementation issues of visually controlled mobile robots
Issues related to the use of different platforms (ground, aerial,
underwater, humanoids, etc.)
Applications of visually controlled mobile robots
Before submission authors should carefully read over the journal’s
Author Guidelines, which are located at
http://www.hindawi.com/journals/jece/guidelines/. Prospective authors
should submit an electronic copy of their complete manuscript through
the journal Manuscript Tracking System at http://mts.hindawi.com/
according to the following timetable:
Manuscript Due Friday, 25 January 2013
First Round of Reviews Friday, 19 April 2013
Publication Date Friday, 14 June 2013
Lead Guest Editor
Mohammed Al-Mualla, Khalifa University of Science, Technology and
Research, P.O. Box 127788, Abu Dhabi, UAE
Jorge Dias, Institute of Systems and Robotics, University of Coimbra,
3030-260 Coimbra, Portugal
Lakmal Seneviratne, Centre for Robotics Research, Department of
Informatics, King’s College London, Strand, London WC2R 2LS, UK
PhD Position in Machine Intelligence
Queen Mary, University of London
School of Electronic Engineering & Computer Science
Applications are invited for a fully funded PhD Studentship starting in September 2013.
The focus of this doctoral research project is on lifelong machine learning. Traditional machine learning methods learn each new problem from scratch, requiring extensive training each time. In contrast, humans rapidly learn to solve new and complex problems with limited practice by building on a lifetime of experience with related tasks and domains. The goal of this project is to develop models for lifelong machine learning, enabling experience from each encountered task and domain to be accumulated and exploited in the next. A variety of applications can be considered as lifelong machine learning impacts diverse areas including computer vision, security, forensics, medical diagnosis, big data, ecommerce and others. Fluency in English and a strong foundation in mathematics (linear algebra, calculus and statistics) and programming are essential. Machine learning and/or computer vision experience is a plus.
Nationality: Open to all.
Deadline: 31 Jan 2013.
Interviews: Feb 2013.
Start: Sep 2013.
For queries contact: Dr. Timothy Hospedales firstname.lastname@example.org.