Techn. Fakultät Willkommen am Institut für Informatik FAU-Logo

Computer Vision

The computer vision group deals with general problems of detecting structures in images. Particular topics currently include color & reflectance, image forensics, multispectral imaging, multi-camera setups and range imaging. Our work is closely related to other main fields in computer vision, like image segmentation and tracking. Particular topics like image forensics connect closely to statistics, color & reflectance serves often as a pre-processing step for higher level computer vision tasks like object recognition.

Colloquium Time Table

Responsible PersonProgram / Topic
21.08.17Marc"Spatial Transformer Networks"
28.08.17Christian J."Deep Unsupervised Similarity Learning using Partially Ordered Sets"
04.09.17VincentCVPR'17 Video Session
11.09.17-Student presentation: Maike Stöve "Artifact Detection in CLE-Images"
18.09.17Peter"UltraStereo: Efficient Learning-Based Matching for Active Stereo Systems"
25.09.17Sergiu"KillingFusion: Non-rigid 3D Reconstruction without Correspondences"
02.10.17VincentCVPR'17 Video Session pt.2
09.10.17Amir"Learning from simulated and unsupervised images through adversarial training"
16.10.17-Student presentation: Maryam Mohammadzahrai "Can classiffers reliably mark dead pixels on real life video sequences?"


-Student presentation: Lara Steffes
6.10.17 Patrick"Sensor Pattern Noise Estimation Based on Improved Locally Adaptive DCT Filtering and Weighted Averaging for Source Camera Identification and Verification"
13.11.17 Benjamin"Source camera identification using auto-white balance approximation"
20.11.17 -Slide session
27.11.17 -no colloquium
4.12.17 VincentStudent presentation (15min) + ICDAR'2017 review
11.12.17 FranzPointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space
18.12.17 MarcSqueeze-and-Excitation Networks
25.12.17 -no colloquium
1.1.18 -no colloquium
8.1.18 -no colloquium
15.1.18 Christian J.Experimental detection of forming limit strains on samples with multiple local necks
22.1.18 -no colloquium
29.1.18 SebastianDiscussion on the organization of the colloquium.
"FlowNet 2.0 Evolution of Optical Flow Estimation with Deep Networks"
5.2.18 -Student presentation: Maryam Mohammadzahrai "Can Classifiers Reliably Mark Dead Pixels on Real Life Video Sequences?"
26.3.18Student presentation: Hooman Mirzaalian "Detecting and measuring surface area of skin lesions"
9.4.18Slide Session + Intro talk Martin Seipert
16.4.18Slide Session Pt 2
"S3FD: Single Shot Scale-Invariant Face Detector"
30.4.18no colloquium

Student short presentation: Oguz Kedilioglu "Real-Time Human Body Localization for Autonomous Multicopters in Industrial Environments" + Paper presentation (Escape from Cells: Deep Kd-Networks for the Recognition of 3D Point Cloud Models)

14.5.18MarcUnpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
28.5.18Christian J.A deep learning approach to estimate stress distribution: a fast and accurate surrogate of finite-element analysis
4.6.18VincentIn Defense of the Triplet Loss for Person Re-Identification

Video Session: Bayesian Deep Learning

18.6.18AlineDetecting People in Artwork with CNNs
25.6.18Franz"Learning to Find Good Correspondences" by Kwang Moo Yi et al.
2.7.18no colloq (Theresa Götz's phd (physic) defense)
9.7.18-Student presentation. Martin Leipert: "Konzeption und Realisierung einer flexiblen 3D-Visualisierung für eine Software zur Simulation medizinischer Robotiksysteme"
16.7.18no colloq
23.7.18AnguelosAbout the DL information bottleneck theory

Mathis: Techniques for balancing unbalanced multi-label datasets.
Sebastian Arndt: BT intro presentation (supervisor: Marc)


Hoffer et al. "Train longer, generalize better: closing the generalization gap in large batch training of neural networks"

15.10.18AnguelosGupta et al. "Learning to Read by Spelling"
22.10.18MarcLin et al. "Focal Loss for Dense Object Detection"
29.10.18Christian J

Brock et al. "Large Scale GAN Training for High Fidelity Natural Image Synthesis"

5.11.18MT Final presentatonOguz, Kedilioglu:  Real-Time Human Body Localization for Autonomous Multicopters in Industrial Environments
12.11.18no colloquium
19.11.18BT Final presentationSebastian Arndt, Detection and Classification of Pulmonary Hemosiderophages in Cytology Slides
26.11.18MT Final presentationSpranger, Lukas: "Learning Global Pooling Operators in Deep Neural Networks for Image Retrieval"
3.12.18no colloquium
10.12.18VincentShama et al. "Adversarial Feedback Loop"
17.12.18MT Final presentationOguz, Kedilioglu:  Real-Time Human Body Localization for Autonomous Multicopters in Industrial Environments

Mailing list subscription management page for Opens external link in new windowstudents and Opens external link in new windowguests.

Running Projects

Historical Document Analysis

Analysis of papal charters from the high middle ages.

Writer Identification

Scribe identification of handwritten documents.

Old Master Drawing Image Analysis

Separating the ink layers of the old master drawings via hyperspectral image Analysis

Image Forensics

Finding forgeries in digital images.

Camera Calibration

Camera calibration aims at obtaining a parametric model of a camera.

Cell Image Processing

Processing of medical image data acquired with microscopes (digital histology) or confocal laser endomicroscopy.

Hyperspectral Remote Sensing Image Analysis

Analysis and Classification of the Hyperspepctral Remote Sensing Images

Sheet Metal Forming Limits

Processing of industrial image data acquired with a stereo camera setup

Finished Projects

Multispectral Image Analysis

Interpreting multispectral data to better understand reflectance phenomena.

Advanced Driver Assistance Systems

Detection and tracking of objects in the vehicle's surrounding.

Solar Irradiance Forecasting

Applying Computer Vision to establish irradiance/power forecasts for solar power plants.

Sensor Fusion for Banknote Processing

Optimizing the sorting results for banknotes through the use of multiple sensor cues.

Single Shot Structured Light

Optical 3D surface measurements with a single image capture.


Student project: small-size autonomous robot soccer.

Skin Reflectance Modeling

Modeling skin reflectance for skin segmentation and illumination color estimation.

Color and Reflectance

Transferring color constancy insights from laboratory setups to real-world images.