Depth Assisted Adaptive Workload Balancing for Parallel View Synthesis

Jin, Xian-Min; Liu, Zhanqi; Li, Qian; Dai, Qionghai

doi:10.1109/tmm.2018.2827781

Cited by 4 publications

(2 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Workload balancing can classify user access into two types: Centralized system and Distributed system [8]. Centralized system is easier to manage in that all systems are concentrated to a specific leader or site [9]. But the disadvantage is that all the management burden is centralized so the server that is responsible for controlling the workload balancing has to work hard in the case of many users [10].…”

Section: Introductionmentioning

confidence: 99%

Efficient Load Balancing for Cloud Computing by Using Content Analysis

Somwang

2022

Int. j. commun. netw. inf. secur.

View full text Add to dashboard Cite

Nowadays, computer networks have grown rapidly due to the demand for information technology management and facilitation of greater functionality. The service provided based on a single machine cannot accommodate large databases. Therefore, single servers must be combined for server group services. The problem in grouping server service is that it is very hard to manage many devices which have different hardware. Cloud computing is an extensive scalable computing infrastructure that shares existing resources. It is a popular option for people and businesses for a number of reasons including cost savings and security. This paper aimed to propose an efficient technique of load balance control by using HA Proxy in cloud computing with the objective of receiving and distributing the workload to the computer server to share the processing resources. The proposed technique applied round-robin scheduling for an efficient resource management of the cloud storage systems that focused on an effective workload balancing and a dynamic replication strategy. The evaluation approach was based on the benchmark data from requests per second and failed requests. The results showed that the proposed technique could improve performance of load balancing by 1,000 request /6.31 sec in cloud computing and generate fewer false alarms

show abstract

Section: Introductionmentioning

confidence: 99%

Efficient Load Balancing for Cloud Computing by Using Content Analysis

Somwang

2022

Int. j. commun. netw. inf. secur.

View full text Add to dashboard Cite

show abstract

“…Thus, the space of the possible view synthesis functions can be extremely large, making it hard to find a good solution. Several recent works [6], [7], [8] attempt to solve the view synthesis problem by warping with a depth camera. However, depth images are difficult to obtain due to the limitation of the hardware.…”

Section: Introductionmentioning

confidence: 99%

Deep View Synthesis via Self-Consistent Generative Network

Liu

Jia

Yang

et al. 2021

Preprint

View full text Add to dashboard Cite

View synthesis aims to produce unseen views from a set of views captured by two or more cameras at different positions. This task is non-trivial since it is hard to conduct pixel-level matching among different views. To address this issue, most existing methods seek to exploit the geometric information to match pixels. However, when the distinct cameras have a large baseline (i.e., far away from each other), severe geometry distortion issues would occur and the geometric information may fail to provide useful guidance, resulting in very blurry synthesized images. To address the above issues, in this paper, we propose a novel deep generative model, called Self-Consistent Generative Network (SCGN), which synthesizes novel views from the given input views without explicitly exploiting the geometric information. The proposed SCGN model consists of two main components, i.e., a View Synthesis Network (VSN) and a View Decomposition Network (VDN), both employing an Encoder-Decoder structure. Here, the VDN seeks to reconstruct input views from the synthesized novel view to preserve the consistency of view synthesis. Thanks to VDN, SCGN is able to synthesize novel views without using any geometric rectification before encoding, making it easier for both training and applications. Finally, adversarial loss is introduced to improve the photo-realism of novel views. Both qualitative and quantitative comparisons against several state-of-the-art methods on two benchmark tasks demonstrated the superiority of our approach.

show abstract