2017
DOI: 10.1016/j.patrec.2016.09.004
|View full text |Cite
|
Sign up to set email alerts
|

Modeling depth for nonparametric foreground segmentation using RGBD devices

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
28
0

Year Published

2017
2017
2021
2021

Publication Types

Select...
4
3
2

Relationship

0
9

Authors

Journals

citations
Cited by 31 publications
(28 citation statements)
references
References 13 publications
0
28
0
Order By: Relevance
“…40,41 Depth data are usually noisy and have some restrictions for certain surfaces in measurement which typically is referred to as "holes" 31 or "Absent Depth Observations (ADO)" in the literature. 11 These failures come from several physical phenomena such as the production of depth camouflage, depth shadows, absorption by black objects, limitation on distances, lower sensitivity at longer distances and absent observations, etc. Figure 1 illustrates the amount of possible noise in each depth frame, for example the black speaker absorbs the signal and consequently the area is defined as absent of observation (shown by black points) or in some part of the cavity, depth is not available due to the characteristics of the scene.…”
Section: Motion Detectionmentioning
confidence: 99%
“…40,41 Depth data are usually noisy and have some restrictions for certain surfaces in measurement which typically is referred to as "holes" 31 or "Absent Depth Observations (ADO)" in the literature. 11 These failures come from several physical phenomena such as the production of depth camouflage, depth shadows, absorption by black objects, limitation on distances, lower sensitivity at longer distances and absent observations, etc. Figure 1 illustrates the amount of possible noise in each depth frame, for example the black speaker absorbs the signal and consequently the area is defined as absent of observation (shown by black points) or in some part of the cavity, depth is not available due to the characteristics of the scene.…”
Section: Motion Detectionmentioning
confidence: 99%
“…The system stores the first N (N = 20 in our experiments) number of frames to initialise the models (system initialization step). Typically, depth frames are noisy and have some limitations for certain materials, surfaces and black color which known as "holes" [17] or "Absent Depth Observations (ADO) " [3]. In the proposed method these unknown pixel values in the depth frame will be filled by neighboring values before storing in the model.…”
Section: The Proposed Algorithmsmentioning
confidence: 99%
“…One of the most common approaches is to compare the current image with previous images which are known as "reference" in the literature. Generally these references are created from a single image or more compound model which is known as a "scene model" [3]. Traditionally a scene model requires a regular update to adapt to the changes over the time in the real-world scenarios.…”
Section: Introductionmentioning
confidence: 99%
“…KDE was used to combine color and depth information for background subtraction after preprocessing. The recent dataset proposed by Gabrilel Moya et al [18] is a new RGB-D dataset for foreground segmentation and generic scene modeling method was designed to analyze the different situations in depth images. Absent Depth Observations (ADO) are divided into two types: caused by scene's physical configuration and resulted from foreground objects edges.…”
Section: A Background Subtraction With Depth Sensorsmentioning
confidence: 99%