Segmentation and tracking of static and moving objects in video surveillance scenarios

Gallego, Jaime; Pardàs, Montse; Landabaso, J.L.

doi:10.1109/icip.2008.4712355

Cited by 19 publications

(19 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some of these approaches [51] directly adopt this method, whereas other strategies complement it with additional stages to improve the FG detection in some situations. In [46] the RGA modeling is combined with Markov Random Fields (MRFs), which spatially smooths the BG modeling results, thus avoiding the need of applying an independent threshold to each pixel and improving the quality of the results in sequences with camera jitter.…”

Section: Single Gaussian Models (Sgm)mentioning

confidence: 97%

See 1 more Smart Citation

Detection of stationary foreground objects: A survey

Cuevas

Martínez

García

2016

Computer Vision and Image Understanding

View full text Add to dashboard Cite

Section: Single Gaussian Models (Sgm)mentioning

confidence: 97%

“…In [51], the SFOs are modeled using a Gaussian distribution, which improves the quality of the results when the color features of the SFOs change slightly. In [92], an edge-based tracking is applied to remove false FG detections (mainly due to illumination changes), ROs and GRs.…”

Section: Persistencementioning

confidence: 99%

Detection of stationary foreground objects: A survey

Cuevas

Martínez

García

2016

Computer Vision and Image Understanding

View full text Add to dashboard Cite

“…Colour and depth images are analysed to detect heads (oval areas of the same depth) [25] and faces by using a modified Viola-Jones detector [26]. Active users are automatically detected and segmented from the background (walls, sofas or chairs) [27]. Once heads are detected, hands are tracked using a 3D virtual box in front of the head of the user with control of the system.…”

Section: Gesture Interactionmentioning

confidence: 99%

Towards a format-agnostic approach for production, delivery and rendering of immersive media

Niamut

Kochale

Ruiz-Hidalgo

et al. 2013

Proceedings of the 4th ACM Multimedia Systems Conference

View full text Add to dashboard Cite

The media industry is currently being pulled in the often-opposing\ud directions of increased realism (high resolution, stereoscopic,\ud large screen) and personalization (selection and control of\ud content, availability on many devices). We investigate the\ud feasibility of an end-to-end format-agnostic approach to support\ud both these trends. In this paper, different aspects of a format-\ud agnostic capture, production, delivery and rendering system are\ud discussed. At the capture stage, the concept of layered scene\ud representation is introduced, including panoramic video and 3D\ud audio capture. At the analysis stage, a virtual director component\ud is discussed that allows for automatic execution of\ud cinematographic principles, using feature tracking and saliency\ud detection. At the delivery stage, resolution-independent\ud audiovisual transport mechanisms for both managed and\ud unmanaged networks are treated. In the rendering stage, a\ud rendering process that includes the manipulation of audiovisual\ud content to match the connected display and loudspeaker properties\ud is introduced. Different parts of the complete system are revisited\ud demonstrating the requirements and the potential of this advanced\ud concept.Peer ReviewedPostprint (published version

show abstract

“…The statistical approaches use the characteristics of individual pixels or groups of pixels to construct more advanced background models [8]. And the statistics of the background can be updated dynamically during processing.…”

Section: Related Workmentioning

confidence: 99%

Object Segmentation from Surveillance Video Sequences

Girisha¹,

Murali²

2010

2010 First International Conference on Integrated Intelligent Computing

View full text Add to dashboard Cite

Identifying moving objects from a video sequence is a fundamental and critical task in many computer vision applications. We propose a three stage adaptive object segmentation algorithm for color surveillance videos. In the first stage, background is modeled using Multiple Correlation Coefficient ( . ) using pixel-level based approach for motion segmentation. Segmented foreground objects generally include their self shadows as foreground objects since the shadow intensity differs and gradually changes from the background in a video sequence. To eliminate such shadows from motion segmented video sequences in the second stage; we propose an algorithm based on inferential statistical Difference in Mean (Z) method. Cast shadows produce troublesome effects for video surveillance systems, typically for object tracking from a fixed viewpoint. To eliminate cast shadows from video sequences in the third stage, we propose an algorithm based on the fact that, cast shadow points are usually adjacent to object points and are merged in a single blob on the edge of the moving objects. The approach uses the Chi-Square Distribution (χ) to build statistical model. χ models are constructed and updated for every inputted frame. Results obtained with different indoor and outdoor sequences show the robustness of the object segmentation approach.

show abstract

Segmentation and tracking of static and moving objects in video surveillance scenarios

Cited by 19 publications

References 9 publications

Detection of stationary foreground objects: A survey

Detection of stationary foreground objects: A survey

Towards a format-agnostic approach for production, delivery and rendering of immersive media

Object Segmentation from Surveillance Video Sequences

Contact Info

Product

Resources

About