Essentials for few-shot learning on images

Project summary

The core principle of few-shot learning is to build on visual feature representations obtained by pre-training neural networks on large image datasets and transfer the learned representation to recognize new object classes from a few samples. We hypothesize that considerable gains can be made by investigating and expanding the foundations of few-shot learning. We aim to provide a concise and solid framework that generalizes to diverse applications. Specifically, we will develop novel core algorithms for few-shot learning, including analysis of the properties of base representations, matching strategies, generative techniques to learn residual features for distinguishing new objects, and techniques for incrementally learning new classes without forgetting previous knowledge.

Our methods

  • Data-driven modeling
  • Neural networks
  • Meta-learning
  • Pre-training
  • Local perspective

Principal investigator 1

Doctoral researcher position 1

(supervised by PI Brox)


Analysis of representation learning techniques for few-shot learning, matching-based few-shot learning, generative methods for few-shot learning.


  • Master’s degree in computer science, mathematics, or physics
  • Strong mathematical background
  • Solid programming experience in Python

Principal investigator 2

Doctoral researcher position 2

(supervised by PI Valada)


Few-shot techniques for learning from 3D data, transductive label propagation, and class-incremental few-shot learning.


  • Master’s degree in computer science, robotics, machine learning, computer vision, or a related field
  • Strong mathematical background
  • Programming skills including in deep learning frameworks such as PyTorch or TensorFlow

Administrative Manager

Marc Schumacher

Institute of Medical Biometry and Statistics,
Faculty of Medicine and Medical Center –
University of Freiburg