Figure 1 - uploaded by P. Chandrashekhariah
Content may be subject to copyright.
Source publication
We introduce a novel active stereo vison-based object tracking system for a humanoid robot. The system tracks a moving object that is dynamically changing its appearance and scale. The system features an in-built learning process that simultaneously learns short term models for the object and potential distractors. These models evolve over time, re...
Contexts in source publication
Context 1
... we present the Hide and Seek system ( Fig. 1), which is an active tracking system imple- mented on a humanoid robot that utilizes stereo vi- sion and segmentation and learns appearance models of objects and potential distractors. It uses these mod- els to explicitly represent difficult situations such as occlusions resulting from the object hiding behind an occluder or being ...
Context 2
... using the distractor dictionary in the same manner as that of object recognition. However, the features will vote for their absolute locations rather than for the centroid. The presence and location of the distractor is decided by merely calculating the density of the distractor's features that are overlap- ping/masking the tracked object (see Fig. ...
Context 3
... system goes into different states as shown in Fig. 11 depending on inputs received from tracker and recognition (object and distraction). This will further guide the decision module of the architecture (Fig. ...
Context 4
... conduct our experiment on the iCub robot head in a regular office room with cluttered background (see Fig. 12). We collect various objects such as a coffee cup, tea packet, coke tin, books etc that have sufficient texture on them. We ask a subject unfamiliar with the system to test the algorithm while we count the success and failure rates for various ...
Context 5
... vision based tracking system on the robot. In con- trast to other systems in the literature, our system at- tempts to identify unusual events in the scene and take necessary actions. The system is also generic enough to be applied for applications such as face/pedestrian tracking on static/active cameras (an instance of face tracking is shown in Fig. 13). There is a scope for fur- ther improvement in every part of our system. How- ever, the present system clearly demonstrates how ob- ject and distractor models allow to recognize difficult tracking situations and to respond to them appropri- ately. This brings us one step closer to building track- ing systems with human level ...
Similar publications
Omnidirectional stereo vision systems have been widely used as primary vision sensors in intelligent robot 3D measurement tasks, which require stereo calibration and rectification. Current stereo calibration and rectification methods suffer from complex calculations or a lack of accuracy. This paper establishes a simple and effective equivalency be...
The article presents the idea of a binocular stereo vision system for surface
inspection and discusses the basic characteristics of epipolar geometry and the advantages
and limitations of the methods of representation that use it. The structure of the vision
system, basic optical parameters and the configuration of the equipment are presented. T...
A handheld 3D laser scanning system is proposed for measuring large-sized objects on site. This system is mainly composed of two CCD cameras and a line laser projector, in which the two CCD cameras constitute a binocular stereo vision system to locate the scanner’s position in the fixed workpiece coordinate system online, meanwhile the left CCD cam...
Braking performance plays a vital role for the safety in vehicle operation, so the measurement of braking performance is very important. In this article, an automatic measurement method for braking performance measurement using binocular stereo vision is proposed in order to obtain the vehicle’s braking trajectory and other performance characterist...
