ADEPT: advanced dense prediction

Semantic analysis of natural images at the pixel level

Summer School on Outliers and Uncertainty in Visual Recognition


Most realistic deployments of visual recognition systems require special considerations in order to ensure proper behaviour in cases of anomalous input. Hence, we are interested in approaches for assessing prediction uncertainty beyond the training distribution. Dealing with uncertainty is especially relevant due to important applications in intelligent transportation, robotics and medical diagnostics. This summer schools encourages sharing of knowledge and exchange of ideas in this fascinating field.

When and Where

The summer school will be held on Wednesday, September 21s at UniZg-FER, D building, third floor, room D1.


09:05 - 10:35
Hanno Gottschalk
From Information to Outliers
10:35 - 11:00 coffee break
11:00 - 12:30
Matthias Rottman
EPFL Lausanne
Notions of Uncertainty
12:30 - 13:45 lunch break
13:45 - 15:15
Matthias Rottman
EPFL Lausanne
Methods for Uncertainty Quantification