Detailed Program (2022)

This workshop will be a full day event on October 24 2022. It will be composed of invited talks and talks for all accepted papers.


All times in Israel Summer Time (GMT+3)

09:00 – 10:00 (60min) – Everything as code (David van Valen)
10:00 – 11:00 (60min)– Session 1 – Object Detection:
  • Empirical Evaluation of Deep Learning Approaches for Landmark Detection in Fish Bioimages (Navdeep Kumar, Ratish Raman, Claudia Di Biagio, Zachary Dellacqua, Arianna Martini, Clara Boglione, Marc Muller, Pierre Geurts, Raphael Maree)
  • PointFISH: learning point cloud representations for RNA localization patterns (Arthur Imbert, Florian Mueller, Thomas Walter)
  • Towards Better Guided Attention and Human Knowledge Insertion in Deep Convolutional Neural Networks (Ankit Gupta, Ida-Maria Sintorn)
11:00 – 12:00 (60min) – Image Denoising for improved Downstream Analysis – a bit of history, recent improvements, and hopes for the future … (Florian Jug)
12:00 – 13:00 (60min) – Session 2 – Data Efficiency:
  • Learning with minimal effort: leveraging ISL for segmentation (Thomas Bonte, Maxence Philbert, Emeline Coleno, Edouard Bertrand, Arthur Imbert, Thomas Walter)
  • Comparison of semi-supervised learning methods for High Content Screening quality control (Masud Umar, Cohen Ethan, Bendidi Ihab, Bollot Guillaume, Genovesio Auguste)
  • Characterization of AI Model Configurations For Model Reuse (Peter Bajcsy, Michael Majurski, Thomas E. Cleveland IV, Manuel Carrasco, Walid Keyrouz)
13:00 – 14:00 (60min)– Lunch break
14:00 – 15:00 (60min) – Reconstructing organelles and macro-molecular structures in
isotropic volume electron microscopy of cells and tissues (Stephan Saalfeld)
15:00 – 16:00 (60min) –


Analyzing animal behavior using machine vision and learning

(Kristin Branson)

16:00 – 17:00 (60min) – Session 3 – Denoising and Attribution
  • Towards Structured Noise Models for Unsupervised Denoising (Benjamin Salmon, Alexander Krull)
  • N2V2 – Fixing Noise2Void Checkerboard Artifacts with Modified Sampling Strategies and a Tweaked Network Architecture (Eva Höck, Tim-Oliver Buchholz, Anselm Brachmann, Florian Jug, Alexander Freytag)
  • Discriminative Attribution from Paired Images (Nils Eckstein, Habib Bukhari, Alexander Shakeel Bates, Gregory Jefferis, Jan Funke)

17:00 – 18:00 (60min) – Self-supervised deep learning encodes high-resolution features of protein subcellular localization (Loic Royer)

Previous editions...