PhD Position Computer Vision / Deep Learning: Video Generation

Context

Inria, the French National Institute for computer science and applied mathematics, promotes “scientific excellence for technology transfer and society”. Graduates from the world’s top universities, Inria's 2,700 employees rise to the challenges of digital sciences. With its open, agile model, Inria is able to explore original approaches with its partners in industry and academia and provide an efficient response to the multidisciplinary and application challenges of the digital transformation. Inria is the source of many innovations that add value and create jobs.
Team
The STARS research team combines advanced theory with cutting edge practice focusing on cognitive vision systems.
Team web site :
https://team.inria.fr/stars/

Assignment

The Inria STARS team is seeking for a Ph.D. researcher with strong background in computer vision, deep learning and machine learning.
The candidate is expected to conduct research with the topic “Generating Synthetic Videos for Automated Visual Learning”, supervised by Antitza Dantcheva in the STARS team, with three axes:
controllable and generalizable human-centric video generation, learning from synthetic data (self-supervised and continual learning), and detection of generated videos and deepfakes in a joint generation/detection setting.

Main activities

Context:

Recent advances in computer vision and generative AI have enabled major breakthroughs in image and video understanding, yet modern deep learning models remain critically dependent on massive amounts of annotated data, whose collection, annotation and curation are costly, time-consuming and increasingly constrained by privacy regulations such as GDPR. This dependency is now one of the main bottlenecks to building robust and generalizable visual learning systems. At the same time, generative models (GANs, diffusion models) have reached a level of realism that makes it possible to synthesize human-centric videos with controllable expressions, gaze, motion and interactions. This thesis explores the foundations of synthetic video generation for automated visual learning: designing new learning paradigms in which synthetic data is generated dynamically to enrich and optimize training sets, with a focus on human-centered tasks (facial expressions, body motion, interaction modeling) and on the joint, co-evolutionary study of generation and detection of synthetic media and deepfakes.

Skills

Candidates must hold a Master degree or equivalent in Computer Science or a closely related discipline by the start date.
The candidate must be grounded in the basics of computer vision, have solid mathematical and programming skills.
Preferably in Python, OpenCV, deep learning framework Pytorch or Tensorflow.
The candidate must be committed to scientific research and strong publications.