Microsoft Research Intern - Computer Vision Algorithms 

Taiwan, Taoyuan City 

882499201

Feature matching is a fundamental step for the solution of many geometric computer vision problems, such as SLAM, 3D registration, and image stitching. However, in each of these problems the ultimate goal is the estimation of geometric parameters related to the scene: the camera motion, the transformation between the 3D objects, the homography relating the images. A Bayesian approach should therefore treat the matches as nuisance parameters, and dispense with them accordingly. In this project we'll attack this problem through a unified framework, using fundamental methods for randomized algorithms, and validate the Bayesian solution against RANSAC and other matching algorithms. Two lines of work will be pursued:

• Development of match-free algorithm for the class of geometric computer vision problems discussed above.
• Careful implementation of the algorithm and its benchmarking against competing methods.

• Currently enrolled in a PhD program in Mathematics, Computer Science, or a related STEM field.

Other Requirements

• Research Interns are expected to be physically located in their manager’s Microsoft worksite location for the duration of their internship.
• In addition to the qualifications below, you’ll need to submit a minimum of two reference letters for this position as well as a cover letter and any relevant work or research samples. After you submit your application, a request for letters may be sent to your list of references on your behalf. Note that reference letters cannot be requested until after you have submitted your application, and furthermore, that they might not be automatically requested for all candidates. You may wish to alert your letter writers in advance, so they will be ready to submit your letter.

• Demonstrated knowledge of probability theory and statistics at the advanced graduate level, in particular sampling and Bayesian methods.
• Proficient in Python or C++.
• Demonstrated ability to develop original an research agenda.
• Drive to publish the work in high-quality academic venues.