Tracking multiple rigid symmetric and non-symmetric objects in real-time using depth data

Akkaladevi, Sharath Chandra; Ankerl, Martin; Heindl, Christoph; Pichler, Andreas

doi:10.1109/icra.2016.7487784

Cited by 23 publications

(17 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…As mentioned in the introduction, the current, widely used standard dataset is the synthetic dataset of Choi and Christensen [4], which contains 4 sequences with 4 objects rendered in a texture-less virtual scene. Another available option is the one provided by Akkaladevi et al [11] who captured a single sequence of a scene containing 4 different objects with a Primesense sensor. However, the 3D models are not complete and do not include training data that could be exploited by learning-based methods.…”

Section: Related Workmentioning

confidence: 99%

A Framework for Evaluating 6-DOF Object Trackers

Garon

Laurendeau

Lalonde

2018

Lecture Notes in Computer Science

View full text Add to dashboard Cite

We present a challenging and realistic novel dataset for evaluating 6-DOF object tracking algorithms. Existing datasets show serious limitations-notably, unrealistic synthetic data, or real data with large fiducial markers-preventing the community from obtaining an accurate picture of the state-of-the-art. Using a data acquisition pipeline based on a commercial motion capture system for acquiring accurate ground truth poses of real objects with respect to a Kinect V2 camera, we build a dataset which contains a total of 297 calibrated sequences. They are acquired in three different scenarios to evaluate the performance of trackers: stability, robustness to occlusion and accuracy during challenging interactions between a person and the object. We conduct an extensive study of a deep 6-DOF tracking architecture and determine a set of optimal parameters. We enhance the architecture and the training methodology to train a 6-DOF tracker that can robustly generalize to objects never seen during training, and demonstrate favorable performance compared to previous approaches trained specifically on the objects to track.

show abstract

Section: Related Workmentioning

confidence: 99%

A Framework for Evaluating 6-DOF Object Trackers

Garon

Laurendeau

Lalonde

2018

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…We report the distance of the prediction's center with the ground truth in Fig. 4 for each object, and compare the performance with that of the method of Akkaladevi et al [1]. In addition to pose estimation errors indicated by the curves (black for Akkaladevi et al [1], red for ours), we also report points at which the trackers lose track of the objects and must be reset.…”

Section: Profactor 3d Datasetmentioning

confidence: 99%

“…To display the errors, we use "boxpercentile plots" [12], which illustrate the distribution of errors vertically. known in a real-life scenario (where the tracker would, for example, be initialized by another algorithm [1]), we also provide an analysis of the sensitivity of the tracker to perturbations in initialization and show results in Fig. 8.…”

Section: Robustness To Initializationmentioning

confidence: 99%

“…In particular, our dataset includes sequences in which the occlusion is slowly varied from low to high levels of occlusions (10%-40% of the object occluded). Since the only available tracking dataset [1] provides only an evaluation for fast moving cameras, we hope that a dataset benchmarking robustness to occlusions will spur research in this area. Finally, we provide an extensive comparison of our work with that of Tan et al [29] and Akkaladevi et al [1], which demonstrates the wide applicability of our approach.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep 6-DOF Tracking

Garon

Lalonde

2017

IEEE Trans. Visual. Comput. Graphics

View full text Add to dashboard Cite

Input RGBD sequence Tracked object Fig. 1. From an input RGBD sequence (top), our method leverages a deep neural network to automatically track the 6-DOF pose of an object even under significant clutter and occlusion (bottom). We demonstrate, through extensive experiments on a novel dataset of real objects with known ground truth pose, that our approach outperforms the state of the art both in terms of accuracy and robustness to occlusions.Abstract-We present a temporal 6-DOF tracking method which leverages deep learning to achieve state-of-the-art performance on challenging datasets of real world capture. Our method is both more accurate and more robust to occlusions than the existing best performing approaches while maintaining real-time performance. To assess its efficacy, we evaluate our approach on several challenging RGBD sequences of real objects in a variety of conditions. Notably, we systematically evaluate robustness to occlusions through a series of sequences where the object to be tracked is increasingly occluded. Finally, our approach is purely data-driven and does not require any hand-designed features: robust tracking is automatically learned from data.

show abstract

“…Für Gegenstände mit Textur ist meist das Aussehen ausreichend für eine Verfolgung der Objektbewegung. Zusätzlich können Methoden aber auch die Form oder Informationen aus Tiefenbildern verwenden, siehe oben oder auch [1].…”

Section: Verfolgung Von Objekten Und Posebestimmungunclassified

Roboter lernen mit Gegenständen umzugehen: neue Entwicklungen und Chancen

Vincze

Zillich

Prankl

2017

Elektrotech. Inftech.

View full text Add to dashboard Cite

Für den zukünftigen Einsatz von Robotern fordern Experten aus der Industrie Roboter, die sicher sind und dem Benutzer eine klare Rückmeldung geben, was der Roboter macht und warum. Eine der wichtigsten Funktionen, um dies zu erreichen, ist die zuverlässige Erkennung der Umgebung, um den Roboter mit einem besseren Verstehen der Welt auszustatten. In diesem Artikel geben wir eine Übersicht derzeitiger Entwicklungen.

show abstract

Tracking multiple rigid symmetric and non-symmetric objects in real-time using depth data

Cited by 23 publications

References 13 publications

A Framework for Evaluating 6-DOF Object Trackers

A Framework for Evaluating 6-DOF Object Trackers

Deep 6-DOF Tracking

Roboter lernen mit Gegenständen umzugehen: neue Entwicklungen und Chancen

Contact Info

Product

Resources

About