People	Publications	Research projects	Downloads

 

Predicting intrinsic properties from a single image using deep networks

The Colour in Context Group (CIC) is offering a postdoc position within the Marie Curie Cofund P-Sphere program at the Computer Vision Center (Universitat Autònoma de Barcelona).

Topic description
Natural images contain many effects caused by the interaction between light and the objects in the scene. These effects include, amongst others, shading, shadows, and specularities. In general, computer vision algorithms are hindered by such effects.

Identifying and isolating these effects and also other physical elements of the scene which participate in the generation of these effects, such as 3D shape, depth and the color of the light source, can be very useful to improve the performance of many algorithms. However, the estimation of these intrinsic properties from a single image is a challenging problem which has received much attention in the last years.

This project aims at the definition of deep network architectures that allow to decompose images into some of their photometric properties such as reflectance, lighting, shading (including self-shadows and cast shadows) or specularities jointly with the corresponding 3D shape properties. To achieve this goal, we propose to work on a large image dataset we are building. This dataset is automatically acquired by creating a wide range of variations in the photometric conditions of the scenes. Scenes are composed by a set of objects whose 3D properties are known.

This proposal is related to the research project MILETRANS (Deep study of intrinsic scene properties to minimize lighting effects in vision-based intelligent transportation systems).

Prof. Maria Vanrell will supervise the Fellow with the support of the hosting group.

Terms and conditions
Duration: 3 years
Salary:
- Living allowance of 36.100€/year (gross salary)
- Single relocation payment of 3000€
Other funding (not included in the salary)
- Research activities: up to 3.600€/year
- Mobility activities: up to 3.600€/year

Candidate's profile  
Applicants should hold:

• A PhD in computer vision, machine learning, or closely related field.
• Some background in deep learning and computational color techniques.
• A good record of publications in the top conferences and journals on computer vision and/or machine learning.
• High motivation and ability to work as part of a team and willing to work with PhD students.
• Proficiency in written and spoken English and good communications skills.
  
  

Aplication
Information about the required documents on the application can be found on this link.
The application deadline is December 9th, 2016.

Research contact:
Name: Maria Vanrell
Email: maria.vanrell@uab.cat