Skip to main content

Post-Doc in Image Processing/Computer Vision, U of Saskatchewan

The Department of Computer Science at the University of Saskatchewan seeks a postdoctoral fellow in the area of image processing and computer vision to take a central role in an international research project funded by the Digging into Data (http://www.diggingintodata.org) program. 

The postdoctoral fellow will be part of the Digging into Archaeological Data: Image Search and Markup (DADAISM) project which is comprised of an international group research teams in three countries.  The DADAISM project seeks to improve the searchability of databases of archaeological images in a manner that facilitates the use cases of archaeological researchers.   The project goals of the DADAISM project are:

1. To develop and deploy a mixed-initiative interactive system that combines minimal human annotation with automated processing techniques to transform the tasks of searching for relevant images and relating them to other electronic resources such as grey literature documents.
2. To investigate how image search and text mining techniques can be used to extract information from image captions and grey literature documents (e.g. field reports) for purposes of finding related images to a candidate image, and for labeling images with appropriate content related data.
3. To investigate how image processing techniques can identify salient features from images of archaeological artifacts for extraction of relevant image similarity measures, and how those techniques generalize across heterogeneous datasets.
4. To investigate whether researchers working with such a mixed-initiative interactive system, based on instrumental interactive techniques, can be more efficient and effective in finding images that are relevant to their archaeological research questions
5. To identify what types of information are required to be present in repositories of artifacts to ensure that the DADAISM approach can be successful across heterogeneous datasets.
Qualifications:  The candidate should hold a Ph.D. in Computer Science or closely related field.  Expertise in image feature extraction and/or content-based image retrieval is essential.  Additional expertise in image segmentation is an asset.  Strong written and oral English communication skills and the ability to produce conference/journal publications independently and in a timely fashion are required. 

Main Responsibilities:  The successful candidate will be expected to drive and conduct the research relating to objective #3 of the project and integrate the work with the other aspects of the project.  They will also have to communicate and coordinate frequently with the other research teams and supervise the day-to-day activities of a research assistant who will be hired for the latter 10 months of the project.

Conditions:  Appointment will be for 18 months, beginning June 1, 2014, at an annual salary of $45,000.

How to apply:  Applicants should email an application package, including a cover letter describing their background and how it suits the described postdoctoral position, a full curriculum vitae, and the names and contact information of at least three references (in a single PDF file), to Dr. Mark Eramian (mark.eramian@usask.ca).    Applicants are asked to provide an email address, a phone number and, ideally, a Skype and/or Facetime address, with their contact information.  Applications will be accepted until April 8, 2014 or the position is filled. 


Comments

Popular posts from this blog

CS/Data Science/Digital Hardware option transfers NOW OPEN

The CS Undergraduate Advising Office has opened applications for CS transfers for Fall 2017. The online application is available at  https://oat.uwaterloo.ca/forms   and will be open from Tuesday, December 5 until  Sunday, December 31, 2017, at 11.59 p.m. We're accepting applications for: Transfer from Math to CS - at a minimum, must have taken CS 136 or be taking it in Fall 2017. Transfer from CS to BCS (Data Science) - must have taken STAT 231 or be taking it in Fall 2017. Transfer from CS to CS/Digital Hardware Adding a Joint CS to an out-of-faculty plan Transfer from outside of Math to CS (pending approval from Math) Answers to common questions about the transfer process can be found in the CS FAQs  (particularly #2 and #151).

AI, Internet of Things, Cybersecurity Online Conference - October 14 & 15, 2017

The world’s largest online conference kicks off this fall stronger than ever by assembling the best  industry leaders, disruptive minds, and visionaries. You’ve read their books and applied their work - now  it’s time to ask them questions, talk to them individually, accept their challenges, and get their feedback!  300 speakers interact with you through live talks, Q&A’s, forum and 1-to-1 video calls. Topics range from  ML to cyber intelligence to industrial IoT. Confirmed speakers for October 14&15th: Ian Goodfellow - Research Scientist, Google Brain Hugo Larochelle - Research Scientist, Google Sandy Carter - VP, Amazon Web Services Louis Monier - Head of AI, AirBnB Tim Abels - Director of Server Architecture, Intel Ashok Banerjee - CTO and VP Engineering, Symantec Andreas Mueller - Lecturer in Data Science, Columbia University Roman Yampolskiy - Assistant Professor, University of Louisville Patrick McDaniel - Distinguished Professor, Penn State ...

GSA SPARK 2020

  GSA SPARK Join us in the fight against SPARK!   In GSA Capital’s boardroom sits SPARK — a decades-old blade server, entombed forever in acrylic to be displayed as a proud reminder of GSA’s journey.   SPARK was an invaluable step toward the sophisticated algorithms in use today, but was retired from active trading and destined never to be booted up again.    But last night, through a series of unfortunate errors, the server was reconnected to GSA’s internal network. In the decades spent sitting on the boardroom shelf its algorithms have warped beyond recognition. SPARK now has a conscience of its own...   Compete to win prizes! Compete with other students in a race to shut SPARK down before it finds a way to dominate every electronic trading market in the world.   For 21 days a range of exciting daily prizes will be on offer to those brave enough to take on SPARK in a series of fiendishly difficult challenges. The contestant deemed to have been most i...