Hazim Kemal Ekenel

TL;DR: Experimental results obtained on the CHIL database, which has been collected from different smart rooms, show that benefiting from multi-view video data and handling registration errors reduce the false identification rates significantly.

TL;DR: The Unconstrained Ear Recognition Challenge (UERC) as mentioned in this paper was a group benchmarking effort centered around the problem of person recognition from ear images captured in uncontrolled conditions, where the goal was to assess the performance of existing ear recognition techniques on a challenging large-scale dataset and identify open problems that need to be addressed in the future.

TL;DR: This analysis shows that methods incorporating deep learning models clearly outperform techniques relying solely on hand-crafted descriptors, even though both groups of techniques exhibit similar behavior when it comes to robustness to various covariates, such as presence of occlusions, changes in (head) pose, or variability in image resolution.

TL;DR: This article presented a new dataset for form understanding in noisy scanned documents (FUNSD) that aims at extracting and structuring the textual content of forms, which can be used for various tasks, including text detection, optical character recognition, spatial layout analysis and entity labeling/linking.

TL;DR: This work uses features from multiple cameras and microphones, and process them in a joint particle filter framework, which performs sampled projections of 3D location hypotheses and scores them using features from both audio and video.