Computer Vision: PhD on Weakly supervised learning for scene understanding

Inria

France

October 31, 2022

Description

2022-05293 - Computer Vision: PhD on Weakly supervised learning for scene understanding

Contract type : Fixed-term contract

Level of qualifications required : Graduate degree or equivalent

Fonction : Temporary scientific engineer

Level of experience : Recently graduated

Context

The candidate will be working on Inria Paris site located in the very heart of Paris, next to the main Gare de Lyon train station.

Assignment

We are looking for a PhD in the Computer Vision group, RITS team. The group is currently expanding and has some exciting subjects coming up. We are opening a research study on weakly-supervised learning for scene understanding.

The work is to study unsupervised or weakly-supervised algorithms to solve semantic segmentation or other vision tasks (object detection, depth estimation, etc.) in unseen adverse conditions. The candidate will have to study the literature and propose a new methodology to progress in this direction. Among others: automatic feature clustering, unsupervised contrastive learning, unsupervised style injection are viable research directions the candidate will have to investigate. During this work we will gradually increase complexity by first considering an open domain transfer learning setup A, domain generalization B, and moving towards true unsupervised setup. The research work will rely on existing deep vision architectures (CNNs, Transformers, etc.).

The candidate is expected to have great knowledge of computer vision and deep learning theory, as well as some practical projects / internships experience on the matter. Good coding ability and experience of deep frameworks are prerequisites.

The PhD will be in the context of Inria / Valeo.ai collaboration. At Inria, the candidate will be in the computer vision group of RITS team, which research deals with scene understanding with a focus on adverse weather and self-supervised learning. Also the candidate will be connected to the valeo.ai team where scientists and engineers conduct fundamental AI research for automotive applications.

He will be co-supervised by Raoul de Charette (Inria), Tuan-Hung Vu (Valeo.ai), Andrei Bursuc (Valeo.ai) and Patrick Pérez (Valeo.ai).

A Liu et al., Open Compound Domain Adaptation. CVPR 2020 B Kim et al., WEDGE: Web-Image Assisted Domain Generalization for Semantic Segmentation. AAAI 2021

Main activities

The precise outline of the work will be refined with the candidate.

Skills

Required skills:

  • Great knowledge of Computer Vision and Deep Learning
  • Knowledge of the main deep vision architectures
  • Ability to read and analyse a scientific article
  • Experience of some of the main deep frameworks
  • Great coding ability
  • Prior scientific experience is a plus
  • The intern must be fluent in english
  • Please also ensure that current diplomacy policies allow your venue on the french territory given the current special sanitary conditions.

    Benefits package
  • Subsidised catering service
  • Partially-reimbursed public transport
  • Flexible working hours
  • Sports facilities
  • General Information
  • Theme/Domain : Robotics and Smart environments Statistics (Big data) (BAP E)

  • Town/city : Paris

  • Inria Center : CRI de Paris
  • Starting date : 2022-10-01
  • Duration of contract : 3 years
  • Deadline to apply : 2022-10-31
  • Contacts
  • Inria Team : RITS
  • Recruiter : De Charette Raoul / raoul.de-charette@inria.fr
  • About Inria

    Inria is the French national research institute dedicated to digital science and technology. It employs 2,600 people. Its 200 agile project teams, generally run jointly with academic partners, include more than 3,500 scientists and engineers working to meet the challenges of digital technology, often at the interface with other disciplines. The Institute also employs numerous talents in over forty different professions. 900 research support staff contribute to the preparation and development of scientific and entrepreneurial projects that have a worldwide impact.

    Instruction to apply

    Defence Security : This position is likely to be situated in a restricted area (ZRR), as defined in Decree No. 2011-1425 relating to the protection of national scientific and technical potential (PPST).Authorisation to enter an area is granted by the director of the unit, following a favourable Ministerial decision, as defined in the decree of 3 July 2012 relating to the PPST. An unfavourable Ministerial decision in respect of a position situated in a ZRR would result in the cancellation of the appointment.

    Recruitment Policy : As part of its diversity policy, all Inria positions are accessible to people with disabilities.

    Warning : you must enter your e-mail address in order to save your application to Inria. Applications must be submitted online on the Inria website. Processing of applications sent from other channels is not guaranteed.