Recursive On-Line &lt;formula formulatype="inline"&gt; &lt;tex Notation="TeX"&gt;${(2{\rm D})}^2{\rm PCA}$&lt;/tex&gt;&lt;/formula&gt; and Its Application to Long-Term Background Subtraction

Seo, Ja-Won; Kim, Seong Dae

doi:10.1109/tmm.2014.2353772

Cited by 15 publications

(2 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then, the discriminative common frame of the test frame is computed by taking the projection of the test frame onto the orthonormal vectors associated with the background list. As shown in (26), the absolute difference between the common frame and the discriminative common frame gives us a novel distance metric for foreground detection. Finally, (27) demonstrates the final distance metric as a combination of the three distance metrics (see (27)) .…”

Section: Foreground Extractionmentioning

confidence: 99%

See 1 more Smart Citation

CVABS: moving object segmentation with common vector approach for videos

Işık

Özkan

Gerek

2019

IET Computer Vision

View full text Add to dashboard Cite

Background modelling is a fundamental step for several real-time computer vision applications that requires security systems and monitoring. An accurate background model helps to detect the activity of moving objects in the video. In this work, the authors have developed a new subspace-based background-modelling algorithm using the concept of common vector approach (CVA) with Gram-Schmidt orthogonalisation. Once the background model that involves the common characteristic of different views corresponding to the same scene is acquired, a smart foreground detection and background updating procedure is applied based on dynamic control parameters. A variety of experiments is conducted on different problem types related to dynamic backgrounds. Several types of metrics are utilised as objective measures and the obtained visual results are judged subjectively. It was observed that the proposed method stands successfully for all problem types reported on CDNet2014 dataset by updating the background frames with a self-learning feedback mechanism.

show abstract

Section: Foreground Extractionmentioning

confidence: 99%

“…Also, it eliminates the dependency to huge number of training data. As a frame level strategy, the low‐dimensional subspace of 2D PCA algorithm was periodically on‐line modified and utilised to recover the background frame [26]. Again, the eigen basis associated with 2D PCA was considered for BMs construction and motion detection.…”

Section: Related Workmentioning

confidence: 99%