Keywords: instance segmentation, self-supervised learning, weakly supervised learning, multimodal analysis.
Place: LIFO – Laboratoire d’Informatique Fondamentale d’Orléans, Orléans, France
Starting: (approximative) May 2026, for 18 months.
Salary : depending on experiences.
The postdoc will contribute to the ANR EnACA project, focusing on AI-based multimodal understanding of comics, manga, and related art forms. The work involves designing a generalizable instance segmentation pipeline robust to visual style and domain shifts using deep learning, generative models, and self-supervised learning.
The candidate is requested to have a PhD, with a strong background in Computer Science/Mathematics/Statistics/Computer Vision or relevant fields. And strong experience in:
– Deep learning frameworks (PyTorch, TensorFlow)
– Instance and semantic segmentation (e.g., Detectron2, Mask R-CNN, SAM)
– Generative models (GANs, diffusion models)
– Self-supervised and contrastive learning
To apply :
Your CV with names of two references has to be sent to the following e-mail:
– Vincent Nguyen (vincent.nguyen@univ-orleans.fr)
Full description : https://tinyurl.com/lifo2026-pd
