A Novel Technique to Remove Marked Dynamic Object from Video Based on Reweighted Low Rank Tensor Completion

Hima, C; Baburaj, M; George, Sudhish N.

doi:10.1109/cccs.2018.8586802

Cited by 2 publications

(4 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First known inpainting methods in literature are not suitable for lling large gaps in the image. The other method is works as tensor completion [19]. Patch-based approaches have also been effective in some studies.…”

Section: Inpainting/completion Image/framementioning

confidence: 99%

Removing Unwanted Objects from Image/Frame By Generating Sub-Images Through Generative Adversarial Network

Choupani

Doǧan

2022

Preprint

View full text Add to dashboard Cite

The amazingly developing and advancing technology in every field also shows itself in image/frame processing. Today, computer vision is an important field of study that can be applied to computers with human vision. Generative networks have been used specifically in studies to specific purpose. After the generating networks were published, they were effectively used to obtain new data for various purposes. Basically, we aimed to remove an unwanted object on the picture or frame. With this aim, thought that like human vision, the foresight of the structures that will replace the object planned to be removed has been considered. Accordingly, if an object is in front of a structure/s the object to desired remove it, this possible data to replace the object must have a similar transitivity with the structure/s behind it. To do this, it is aimed to create new data by using sub-data in object location surrounding the object, with the approach that the appropriate data to replace the object can be filled by learning from the structures around the object. It was tried to replace the object (for pixels containing the object) by generating pixels in object dimensions, using generative networks and useful results were obtained, in this study.

show abstract

Section: Inpainting/completion Image/framementioning

confidence: 99%

Removing Unwanted Objects from Image/Frame By Generating Sub-Images Through Generative Adversarial Network

Choupani

Doǧan

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…For a tensor of order τ, the decomposition of the tensor sequence can be defined in the form of formula (6). Among them, ϑ ðkÞ represents the tensor kernel.…”

Section: Approximate Tensor Solution For Rank Of Tensormentioning

confidence: 99%

“…Algorithm 1: Approximate tensor solving algorithm based on tensor column rank. 6 Wireless Communications and Mobile Computing g k . Then, normalize g k based on formula (8) to obtain the following:…”

Section: Flow Of Adaptive Weighting Algorithm Based On Rank Ofmentioning

confidence: 99%

“…A spaRSE prior is introduced to complete the automatic determination of the rank. Hima et al [6] proposed a new low-rank tensor model based on cyclic algebra, that is, the twist Tensor Nuclear Norm (t-TNN) algorithm. Chen and Zhou [7] focused on the method of repairing missing areas in streaming media after removing unwanted moving objects.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Perceived Integrity of Distributed Streaming Media Based on AWTC-TT Algorithm Optimization

Jiang

Zhai

et al. 2022

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

With the development of economy, more and more attention has been paid to the monitoring system, which provides a reliable and powerful guarantee for people’s daily life, property security, and national security. The intelligent video surveillance introduces computer vision-related technologies into traditional video surveillance and realizes the analysis and understanding of video data without artificial dependence to obtain valuable target information in the perceived video data. On this basis, functions such as abnormal event monitoring and real-time alarm are realized. Distributed streaming media monitoring has changed the manual-based monitoring and content analysis modes of traditional monitoring, but the high-complexity calculations such as motion estimation and motion compensation in the encoding process increase the burden of monitoring and sensing equipment. Especially with the development of wireless multimedia technology, the traditional video coding has been unable to meet the requirements of monitoring and sensing equipment in the monitoring system based on wireless technology. This paper proposes an adaptive weighted tensor completion algorithm to complete the repair of streaming media data perceived by ordinary sensing devices. In the proposed algorithm, considering the unbalanced information distribution and data redundancy problems that may exist in the data, the tensor data is adjusted according to the approximate solution algorithm to obtain tensor data that only retains important information and the information distribution is more balanced and reasonable. In the iterative solution process, in order to better map the impact of each dimension of data in the repair process, an adaptive weighting mechanism is proposed according to the data characteristics to obtain the corresponding weight value of each dimension of data. Finally, the proposed approximate tensor solving algorithm and adaptive weighting mechanism are applied to a simple low-rank tensor completeness algorithm based on tensor columns to form the algorithm of this paper, and it is used to repair perceptual streaming media data with data missing problems. The experimental results show that the algorithm in this paper can improve the perceived streaming media data quality by 3% based on the known data information and maintain an advantage of 2% in average processing time. It avoids the replacement of sensing equipment and also provides data quality assurance for subsequent sensing streaming media content analysis. It has certain research significance for the development of monitoring system with artificial intelligence management for target perception and tracking.

show abstract

A Novel Technique to Remove Marked Dynamic Object from Video Based on Reweighted Low Rank Tensor Completion

Cited by 2 publications

References 17 publications

Removing Unwanted Objects from Image/Frame By Generating Sub-Images Through Generative Adversarial Network

Removing Unwanted Objects from Image/Frame By Generating Sub-Images Through Generative Adversarial Network

Perceived Integrity of Distributed Streaming Media Based on AWTC-TT Algorithm Optimization

Contact Info

Product

Resources

About