Martijn Liem

Learn More
The visual tracking of people is an essential capability in many surveillance applications. In this paper, we are interested estimating ground plane location in the more challenging cases involving multiple persons and dynamic environments (e.g. uncontrolled, outdoor settings). To cope with occlusions we use three overlapping, off-line calibrated cameras.(More)
The assignment of multiple person tracks to a set of candidate person locations in overlapping camera views is potentially compu-tationaly intractable, as observables might depend upon visibility order, and thus upon the decision which of the candidate locations represent actual persons and which do not. In this paper, we present an approximate assignment(More)
The capability to follow a person in a domestic environment is an important prerequisite for a robot companion. In this paper, a tracking algorithm is presented that makes it possible to follow a person using a small robot. This algorithm can track a person while moving around, regardless of the sometimes erratic movements of the legged robot. Robust(More)
We present a comparative study for tracking multiple persons using cameras with overlapping views. The evaluated methods consist of two batch mode trackers (Berclaz et al, 2011, Ben-Shitrit et al, 2011) and one recursive tracker (Liem and Gavrila, 2011), which integrate appearance cues and temporal information differently. We also added our own improved(More)
One of the major parts of the multimedia research is object localisation and object tracking. In this paper we try to create a surveillance system that uses object tracking and background subtraction. The final system can keep track of a person walking trough a room and detect if something is left behind. For the tracker, adaptive feature selection is used(More)
— We present a novel approach for the joint estimation of a person's overall body orientation, 3D shape and texture , from overlapping cameras. Overall body orientation (i.e. rotation around torso major axis) is estimated by minimizing the difference between a learned texture model in a canonical orientation and a texture sampled using the current 3D shape(More)
We present a system to track the positions of multiple persons in a scene from overlapping cameras. The distinguishing aspect of our method is a novel, two-step approach that jointly estimates person position and track assignment. The proposed approach keeps solving the assignment problem tractable, while taking into account how different assignments(More)
This paper presents a multi-camera system to track multiple persons in complex, dynamic environments. Position measurements are obtained by carving out the space defined by foreground regions in the overlapping camera views and projecting these onto blobs on the ground plane. Person appearance is described in terms of the colour his-tograms in the various(More)
In order to create a social robot, able to interact with people in a natural way, one of the things it should be able to do is to follow a person around an area. This master's research is aimed at enabling a Sony AIBO robot dog to watch people closely and follow them around the house. This is done using the visual data provided by the robot's nose-mounted(More)