Multi-object detection and tracking by stereo vision

Cai, Ling; Xu, Yingchao; Zhao, Yuming; Yang, Xin

doi:10.1016/j.patcog.2010.06.012

Cited by 44 publications

(20 citation statements)

References 44 publications

(77 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The developments in video acquisition technology in the past decade led to an increasing use of multiple view systems in place of the monocular ones. For example, surveillance systems appeared that consist of one [19] or multiple stereo cameras [20] or multiple single-view cameras. These systems exploit the additional information obtained by exploiting the stereo geometry, namely the disparity information.…”

Section: Related Workmentioning

confidence: 99%

Stereo object tracking with fusion of texture, color and disparity information

Zoidi

Nikolaidis

Tefas

et al. 2014

Signal Processing: Image Communication

View full text Add to dashboard Cite

A novel method for visual object tracking in stereo videos is proposed, which fuses an appearance based representation of the object based on Local Steering Kernel features and 2D color-disparity histogram information. The algorithm employs Kalman filtering for object position prediction and a sampling technique for selecting the candidate object regions of interest in the left and right channels. Disparity information is exploited, for matching corresponding regions in the left and right video frames. As tracking evolves, any significant changes in object appearance due to scale, rotation, or deformation are identified and embodied in the object model. The object appearance changes are identified simultaneously in the left and right channel video frames, ensuring correct 3D representation of the resulting bounding box in a 3D display monitor.The proposed framework performs stereo object tracking and it is suitable for application in 3D movies, 3D TV content and 3D video content captured by consuming stereo cameras. Experimental results proved the effectiveness of the proposed method in tracking objects under geometrical transformations, zooming and partial occlusion, as well as in tracking slowly deforming articulated 3D objects in stereo video.

show abstract

Section: Related Workmentioning

confidence: 99%

Stereo object tracking with fusion of texture, color and disparity information

Zoidi

Nikolaidis

Tefas

et al. 2014

Signal Processing: Image Communication

View full text Add to dashboard Cite

show abstract

“…Some methods based on exploiting feature points, can separate object's points by estimating the background [6]. While the cameras are static, a region of interest (ROI) can be defined to omit the background points which lie out of region [17]. We choose to use the second method of removing the background points since the cameras are motionless and there is no undesired moving object in the region.…”

Section: A Object Detectionmentioning

confidence: 99%

Robust approach for people detection and tracking by stereo vision

Abbaspour

Yazdi

Shirazi

2014

7'th International Symposium on Telecommunications (IST'2014)

View full text Add to dashboard Cite

In this paper, a novel method for people detection and tracking is proposed, based on stereo vision. Each person is represented by a group of the feature points. In this method feature point extraction and 2D space construction of projected points on the ground plane is performed in order to provide top view. Occlusion, as a main challenge in tracking systems, can be addressed by top view scene. A robust kernel density estimation method is employed to categorize points. Then Kalman filter is applied to reduce the detection computation complexity from second frame by predicting center of the groups in the next frame. Our method is more practical than existing methods since it has lower computation cost of detection, because of using feature extraction instead of depth map. This low computational complexity makes our method suitable to be used in real time applications.

show abstract

“…freq(P ) = 5. Since occurrences 1 and 4 are close to each other, i.e., their spatial distance is lower than 2 and their temporal distance is 2 ≤ τ , there is an edge (1,4) in the occurrences graph of P . Conversely, the edges (3, 5) or (2, 11) do not exist in the occurrences graph, as the spatial distance between 3 and 5 or the temporal distance between 2 and 11 are too large.…”

Section: Definition 9 (Frequency Of a Spatio-temporal Pattern) The Fmentioning

confidence: 99%

“…1 occurrences graph(P ) = empty graph 2 for each occurrence of P in D do 3 Add this occurrence to occurrences graph(P ) 4 Computes all valid extensions of this occurrence 5 Computes the edges of occurrences graph(P ) (using and τ ) 6 Computes all spatio-temporal patterns based on P 7 for each spatio-temporal pattern S based on P do 8 if freq st (S) ≥ minfreq st then output(S) 9 if there is no frequent spatio-temporal pattern then return 10 else 11 for each extension E of P do 12 if the code of E ∪ P is canonical and freq(E ∪ P ) ≥ minfreq then 13…”

Section: Dyplagram St Algorithmmentioning

confidence: 99%

“…When it is the case as in [4], multiple cameras are used to tackle object occlusions or features instability using stereo vision. The setting taken in [8] is close to the one we are interested in since they consider cameras embedded in surveillance cars but they rely on strong background information (here GPS position) to perform an effective tracking.…”

Section: Introduction and Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Graph Mining for Object Tracking in Videos

Diot

Fromont

Jeudy

et al. 2012

Machine Learning and Knowledge Discovery in Databases

View full text Add to dashboard Cite

Abstract. This paper shows a concrete example of the use of graph mining for tracking objects in videos with moving cameras and without any contextual information on the objects to track. To make the mining algorithm efficient, we benefit from a video representation based on dynamic (evolving through time) planar graphs. We then define a number of constraints to efficiently find our so-called spatio-temporal graph patterns. Those patterns are linked through an occurrences graph to allow us to tackle occlusion or graph features instability problems in the video. Experiments on synthetic and real videos show that our method is effective and allows us to find relevant patterns for our tracking application.

show abstract

Multi-object detection and tracking by stereo vision

Cited by 44 publications

References 44 publications

Stereo object tracking with fusion of texture, color and disparity information

Stereo object tracking with fusion of texture, color and disparity information

Robust approach for people detection and tracking by stereo vision

Graph Mining for Object Tracking in Videos

Contact Info

Product

Resources

About