direkt zum Inhalt springen

direkt zum Hauptnavigationsmenü

Sie sind hier

TU Berlin

Inhalt des Dokuments

Forschung im Fachbereich Nachrichtenübertragung

Liste der geförderte Forschungsprojekte
Aktuelle Forschungsgebiete

IOU Tracker

Tracking-by-detection is a common approach to multi-object tracking. With ever increasing performances of object detectors, the basis for a tracker becomes much more reliable. In combination with commonly higher frame rates, this poses a shift in the challenges for a successful tracker. We propose a very simple tracking algorithm which can compete with more sophisticated approaches at a fraction of the computational cost. With thorough experiments we show its potential using a wide range of object detectors. The proposed method can easily run at thousands of frames per second (fps) while outperforming the state-of-the-art on the DETRAC vehicle tracking dataset and achieves competitive results on the MOT17 benchmark. A python implementation of the tracker is publicly available. ...


High Efficiency Video Coding

Coding artifacts in video codecs can be reduced using several spatial in-loop filters which are part of the emerging video coding standard HEVC. In this paper, we introduce the concept of global motion temporal filtering (GMTF). A theoretical framework for a concept combining the temporal overlapping of several noisy versions of the same signal is introduced. This includes a model of the motion estimation error. As an important result it is shown that an optimum number of frames N for filtering exists. ...


Optical Flow based Motion Estimation

Optical flow based motion estimation methods are important components of nowadays computer vision applications. We focus on the improvement of the accuracy and run-time performances of local optical flow methods. Knowing that in general the accuracy of global optical flow methods are more accurate than local when comparing dense motion fields, we focused on the fields of application that apply sparse motion information e.g. for tracking, video-based surveillance or video coding. ... 



The Probability Hypothesis Density (PHD) filter is a multi-object Bayes filter which has recently attracted a lot of interest in the tracking community mainly for its linear complexity and its ability to deal with high clutter especially in radar/sonar scenarios. In the computer vision community however, underlying constraints are different from radar scenarios and have to be taken into account when using the PHD filter...


Multimodal Geo-Tagging

We present a hierarchical, multi-modal approach for geo-referencing Flickr videos. Our approach makes use of external resources to identify toponyms in the metadata and of visual features to identify similar content. We use a database of more than 3.2 million Flickr images to group them into geographical areas and to build a hierarchical model. First, the geographical boundaries extraction method identifies the country and its dimension.  ...


Consistent Two-Level Metric

Since the commonly used benchmarks for abandoned object detection (AOD) only have few abandoned objects and a non-standardized evaluation procedure, an objective performance comparison between different methods is hard. Therefore, we propose a new evaluation metric which is focused on an end-user application case and an evaluation protocol which eliminates uncertainties in previous performance assessments.

Software & Datasets
Background Substraction / Foreground Detection
SGMM-SOD Library
Robust Local Optical Flow
RLOF Library
Multi-Object and Multi-Camera Tracking Dataset
MOCAT Dataset
Evaluation framework for abandoned object  detection
IOU Tracker
Code on Github

Zusatzinformationen / Extras


Schnellnavigation zur Seite über Nummerneingabe