Datasets

To allow for replication of the results published in most of my papers,
I tried to make as much code available as possible.

Flower Box
... for Obstacle Avoidance and Accessible Section Detection

A dataset for accessible section detection and obstacle avoidance. Recorded with a handheld stereo camera rig subject to strong ego-motion. 20 videos of varying length covering common urban scenes. For the full dataset, use the get-flower-box.sh script (downloads and unpacks all files into 'flower-box' directory) or first download and extract the main flower-box-base.zip file (contains the Readme and metadata, e.g. labels) as well as all the individual video zip files (click on the images or follow the link to the data folder and unzip all files to the 'img' folder).

Google-512
... for Color Learning and Transfer Learning

A dataset for accessible section detection and obstacle avoidance. Recorded with a handheld stereo camera rig subject to strong ego-motion. 20 videos of varying length covering common urban scenes. For the full dataset, use the get-flower-box.sh script (downloads and unpacks all files into 'flower-box' directory) or first download and extract the main flower-box-base.zip file (contains the Readme and metadata, e.g. labels) as well as all the individual video zip files (click on the images or follow the link to the data folder and unzip all files to the 'img' folder).

Face Detection for Moran Cerf's FIFA Dataset
... for Visual Saliency and Eye-Tracking

MCT-based face detection on the Cerf/FIFA eye-tracking data set to model the influence of faces on visual saliency.

Jasper's Multicam
... for Multi-Camera Localization and Calibration

Dataset to evaluate multi-camera localization and iterative self-calibration of a multi-camera network in a smart room.