Novel Gaussian Mixture-based Video Coding for Fixed Background Video Streaming

Ghafari, Mohammadreza; Amirkhani, Abdollah; Ghanbari, Shirin

doi:10.1109/mvip53647.2022.9738789

Cited by 3 publications

(1 citation statement)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To model the background model, statistical (or parametric) methods using Gaussians are proposed [20,25,26] to label each pixel as a background or foreground pixel. In [27], an improved Gaussian mixture model compression technique is proposed to separate background images and moving objects in video. However, parametric methods are computationally inefficient.…”

Section: Conventional Methodsmentioning

confidence: 99%

Real-Time Motion Detection Network Based on Single Linear Bottleneck and Pooling Compensation

2022

View full text Add to dashboard Cite

Motion (change) detection is a basic preprocessing step in video processing, which has many application scenarios. One challenge is that deep learning-based methods require high computation power to improve their accuracy. In this paper, we introduce a novel semantic segmentation and lightweight-based network for motion detection, called Real-time Motion Detection Network Based on Single Linear Bottleneck and Pooling Compensation (MDNet-LBPC). In the feature extraction stage, the most computationally expensive CNN block is replaced with our single linear bottleneck operator to reduce the computational cost. During the decoder stage, our pooling compensation mechanism can supplement the useful motion detection information. To our best knowledge, this is the first work to use the lightweight operator to solve the motion detection task. We show that the acceleration performance of the single linear bottleneck is 5% higher than that of the linear bottleneck, which is more suitable for improving the efficiency of model inference. On the dataset CDNet2014, MDNet-LBPC increases the frames per second (FPS) metric by 123 compared to the suboptimal method FgSegNet_v2, ranking first in inference speed. Meanwhile, our MDNet-LBPC achieves 95.74% on the accuracy metric, which is comparable to the state-of-the-art methods.

show abstract

Section: Conventional Methodsmentioning

confidence: 99%

Real-Time Motion Detection Network Based on Single Linear Bottleneck and Pooling Compensation

2022

View full text Add to dashboard Cite

show abstract

Image-Feature Parallel Compression for Indoor Surveillance Video

Meng,

Wang,

Yin

et al. 2023

2023 IEEE 25th International Workshop on Multimedia Signal Processing (MMSP)

View full text Add to dashboard Cite

AI Based Video Processing using OO

Pavishna A M,

Deeksha R,

Nikitha G

et al. 2023

IJARSCT

View full text Add to dashboard Cite

The increase in digital video material has increased the requirement for efficient video processing methods. Conventional techniques of video processing are sometimes time-consuming and reliant on substantial user intervention, resulting in decreased efficiency and precision. Video processing based on Artificial Intelligence (AI) is developing as a feasible alternative to conventional approaches. In this study, we investigate the application of Object-Oriented AI concepts to video processing. Our method utilizes object identification and recognition algorithms based on deep learning to recognize and track objects in video streams. In addition, we illustrate the efficacy of our method by applying it to a variety of actual video processing settings

show abstract

Novel Gaussian Mixture-based Video Coding for Fixed Background Video Streaming

Cited by 3 publications

References 15 publications

Real-Time Motion Detection Network Based on Single Linear Bottleneck and Pooling Compensation

Real-Time Motion Detection Network Based on Single Linear Bottleneck and Pooling Compensation

Image-Feature Parallel Compression for Indoor Surveillance Video

AI Based Video Processing using OO

Contact Info

Product

Resources

About