• Home
• Group
  • Staff
  • Finding Us
  • Software
  • Pictures
  • Internal
• News
  • Press Releases
  • Information
  • Dates
• Lectures
  • Rechnersehen 1
  • Rechnersehen 2
  • Bildinformatik
  • Zustandsschätzung
  • Tiefe Lernverfahren
  • Visuelle Objekterkennung
  • Bewegungsberechnung
  • Signal. Bildverarbeitung
  • Spezielle Probleme
  • Seminar Rechnersehen
  • Anwendungspraktikum
  • Image Processing
  • Informatik (B.Sc. Werkstoffwissenschaften)
  • Informatik I (B.Sc. Physik)
  • Informatik II (B.Sc. Physik)
  • Programmieren in C++ (B.Sc. Mathematik)
  • Objektorientierte Programmierung mit C++ (ASQ)
  • Online-Lehre im SoSe 21
  • Final Theses
  • References
• Research
  • Lifelong Learning
  • Image Understanding
  • Learning from 3D and Unstructured Data
  • AMMOD
  • Finished Projects
  • Datasets
  • CNN Models
  • Understanding Deep Learning
• Publications
• Jobs and Applications
  • Open positions of the 6th regular iDiv Flexpool call
• Cooperations

Publication in Journal of Imaging

A study conducted by our group about the presence and effect of duplicate images in popular computer vision benchmark datasets has been published in the MDPI Journal of Imaging!

• Do We Train on Test Data? Purging CIFAR of Near-Duplicates by Björn Barz and Joachim Denzler

The paper is available as free of charge as open access publication.

Further information can also be found on our project page.