Probability occupancy maps for occluded depth images

Bagautdinov, Timur; Fleuret, François; Fua, Pascal

doi:10.1109/cvpr.2015.7298900

Cited by 42 publications

(41 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…4. Without occlusion tracking as in [14] and [15], the identity of occluded objects cannot be retained after the occlusion. Inter-sensor identification depends on the pre-defined settings for determining the overlapping area.…”

Section: Resultsmentioning

confidence: 99%

“…and correctly identify the object even when the object is occluded with moving subjects (such as moving hands or books). Object detection and tracking system without occlusion handling using depth-only image was developed in [13], while detection and tracking of multiple objects with occlusion detection but no identification during the occlusion were presented in [14] and [15]. Our proposed depth-based object detection and tracking method enables individual object identification for each object during occlusion.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Two-layer tracking for occlusion handling and inter-sensor identification in multiple depth sensorsbased object detection and tracking

Sabirin

Naito

2017

2017 25th European Signal Processing Conference (EUSIPCO)

View full text Add to dashboard Cite

Abstract-The main challenge in depth-based object detection and tracking process is to provide correct identification of the detected objects during occlusion. This is because the information necessary to distinguish and consequently identify the objects throughout the occlusion events are limited, compared to conventional, color-based object tracking. In this paper we propose a two-layer tracking method that enables automatic occlusion handling and inter-sensor identification for object detection and tracking that utilizes more than one depth sensor. On the first layer, the tracking is first performed independently for each sensor to extract objects' feature and perform initial tracking with separation of the occluded objects. On the second layer, the tracking is performed in the perspective projection of the objects tracked on the first layer that are combined in a single processing plane to provide correct identification of the objects that are detected in one sensor to another. Experiment results show that the proposed method can correctly identified occluded objects and objects that are moving between sensors coverage area.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Two-layer tracking for occlusion handling and inter-sensor identification in multiple depth sensorsbased object detection and tracking

Sabirin

Naito

2017

2017 25th European Signal Processing Conference (EUSIPCO)

View full text Add to dashboard Cite

show abstract

“…Bagautdinov et al . [BFF15] identify objects, especially persons, standing on the floor of indoor rooms acquired through depth sensors. The algorithm builds a Bayesian generative model of probabilistic occupancies at each location.…”

Section: Theoretical Foundationsmentioning

confidence: 99%

A Survey of Simple Geometric Primitives Detection Methods for Captured 3D Data

Kaiser

Zepeda

Boubekeur

2018

Computer Graphics Forum

103

View full text Add to dashboard Cite

The amount of captured 3D data is continuously increasing, with the democratization of consumer depth cameras, the development of modern multi‐view stereo capture setups and the rise of single‐view 3D capture based on machine learning. The analysis and representation of this ever growing volume of 3D data, often corrupted with acquisition noise and reconstruction artefacts, is a serious challenge at the frontier between computer graphics and computer vision. To that end, segmentation and optimization are crucial analysis components of the shape abstraction process, which can themselves be greatly simplified when performed on lightened geometric formats. In this survey, we review the algorithms which extract simple geometric primitives from raw dense 3D data. After giving an introduction to these techniques, from the acquisition modality to the underlying theoretical concepts, we propose an application‐oriented characterization, designed to help select an appropriate method based on one's application needs and compare recent approaches. We conclude by giving hints for how to evaluate these methods and a set of research challenges to be explored.

show abstract

“…Derinlik algılayıcılarıyla bugüne kadar yapılmış insan takibi amaçlı çalışmaların çogunda, insanların ayakta durdukları ya da yürüdükleri varsayılmıştır [1], [2], [3], [4], [5]. Bu çalışmalarda, deneyler, laboratuar ya da okul koridoru gibi tek bir mekanda gerçekleştirilmiştir [1], [2], [5], [6].…”

Section: Introductionunclassified

“…Bu çalışmalarda, deneyler, laboratuar ya da okul koridoru gibi tek bir mekanda gerçekleştirilmiştir [1], [2], [5], [6]. Oysa, insanların çok çeşitli pozlarda, farklı karmaşık ortamlarda, hareket bilgisine dayanmadan sezimi, yaşlı ve hasta takibi, felaket sonrası iç ortamlarda arama kurtarma gibi uygulamalarda, servis robotlarının görevlerini yerine getirmelerinde önemli bir aşamadır.…”

Section: Introductionunclassified

Detection of humans from depth images

Can

Çakır

Namli

et al. 2016

2016 24th Signal Processing and Communication Application Conference (SIU)

View full text Add to dashboard Cite

Özetçe -Derinlik algılayıcılarının tüketici kullanımına uygun hale gelmesiyle birlikte, derinlik imgelerinden insan sezimi ve takibi üzerindeki araştırmalar hız kazanmıştır. Derinlik bilgisi, nesnelerin arkaplandan ayrılmasını kolaylaştırmakta, ayrıca bulunan nesnelerin 3B uzayda konumlandırılmasına olanak saglamaktadır. Bu çalışmada, ev, ofis, kafe gibi iç ortamlardan alınmış, insanların çok çeşitli pozlarda bulundugu derinlik imgelerinden oluşan yeni bir veritabanı sunulmaktadır. Hareketsiz insanların sezimi için yeni bir yöntem önerilmiş, bu yöntem yeni veritabanımız üzerinde denenmiştir. 1 Anahtar Kelimeler-insan sezimi, derinlik imgesi, Microsoft KinectAbstract-With the development of consumer depth sensors, research on human detection and tracking from depth images has gained momentum. Depth information facilitates the extraction of objects from the background, and enables localization of these objects in 3D space. In this work, we present a new dataset of depth images acquired from indoor environments, such as home, office, coffee shop, where people are present in a variety of poses. We propose a new method for detection of unmoving humans, and test our algorithm on our new dataset.

show abstract

Probability occupancy maps for occluded depth images

Cited by 42 publications

References 18 publications

Two-layer tracking for occlusion handling and inter-sensor identification in multiple depth sensorsbased object detection and tracking

Two-layer tracking for occlusion handling and inter-sensor identification in multiple depth sensorsbased object detection and tracking

A Survey of Simple Geometric Primitives Detection Methods for Captured 3D Data

Detection of humans from depth images

Contact Info

Product

Resources

About