PIB:用于协作式边缘视频分析的优先信息瓶颈框架

Zhengru Fang, Senkang Hu, Liyan Yang, Yiqin Deng, Xianhao Chen, Yuguang Fang
{"title":"PIB:用于协作式边缘视频分析的优先信息瓶颈框架","authors":"Zhengru Fang, Senkang Hu, Liyan Yang, Yiqin Deng, Xianhao Chen, Yuguang Fang","doi":"arxiv-2408.17047","DOIUrl":null,"url":null,"abstract":"Collaborative edge sensing systems, particularly in collaborative perception\nsystems in autonomous driving, can significantly enhance tracking accuracy and\nreduce blind spots with multi-view sensing capabilities. However, their limited\nchannel capacity and the redundancy in sensory data pose significant\nchallenges, affecting the performance of collaborative inference tasks. To\ntackle these issues, we introduce a Prioritized Information Bottleneck (PIB)\nframework for collaborative edge video analytics. We first propose a\npriority-based inference mechanism that jointly considers the signal-to-noise\nratio (SNR) and the camera's coverage area of the region of interest (RoI). To\nenable efficient inference, PIB reduces video redundancy in both spatial and\ntemporal domains and transmits only the essential information for the\ndownstream inference tasks. This eliminates the need to reconstruct videos on\nthe edge server while maintaining low latency. Specifically, it derives\ncompact, task-relevant features by employing the deterministic information\nbottleneck (IB) method, which strikes a balance between feature informativeness\nand communication costs. Given the computational challenges caused by IB-based\nobjectives with high-dimensional data, we resort to variational approximations\nfor feasible optimization. Compared to TOCOM-TEM, JPEG, and HEVC, PIB achieves\nan improvement of up to 15.1\\% in mean object detection accuracy (MODA) and\nreduces communication costs by 66.7% when edge cameras experience poor channel\nconditions.","PeriodicalId":501280,"journal":{"name":"arXiv - CS - Networking and Internet Architecture","volume":"1 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"PIB: Prioritized Information Bottleneck Framework for Collaborative Edge Video Analytics\",\"authors\":\"Zhengru Fang, Senkang Hu, Liyan Yang, Yiqin Deng, Xianhao Chen, Yuguang Fang\",\"doi\":\"arxiv-2408.17047\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Collaborative edge sensing systems, particularly in collaborative perception\\nsystems in autonomous driving, can significantly enhance tracking accuracy and\\nreduce blind spots with multi-view sensing capabilities. However, their limited\\nchannel capacity and the redundancy in sensory data pose significant\\nchallenges, affecting the performance of collaborative inference tasks. To\\ntackle these issues, we introduce a Prioritized Information Bottleneck (PIB)\\nframework for collaborative edge video analytics. We first propose a\\npriority-based inference mechanism that jointly considers the signal-to-noise\\nratio (SNR) and the camera's coverage area of the region of interest (RoI). To\\nenable efficient inference, PIB reduces video redundancy in both spatial and\\ntemporal domains and transmits only the essential information for the\\ndownstream inference tasks. This eliminates the need to reconstruct videos on\\nthe edge server while maintaining low latency. Specifically, it derives\\ncompact, task-relevant features by employing the deterministic information\\nbottleneck (IB) method, which strikes a balance between feature informativeness\\nand communication costs. Given the computational challenges caused by IB-based\\nobjectives with high-dimensional data, we resort to variational approximations\\nfor feasible optimization. Compared to TOCOM-TEM, JPEG, and HEVC, PIB achieves\\nan improvement of up to 15.1\\\\% in mean object detection accuracy (MODA) and\\nreduces communication costs by 66.7% when edge cameras experience poor channel\\nconditions.\",\"PeriodicalId\":501280,\"journal\":{\"name\":\"arXiv - CS - Networking and Internet Architecture\",\"volume\":\"1 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-08-30\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - CS - Networking and Internet Architecture\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2408.17047\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Networking and Internet Architecture","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2408.17047","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

协作边缘传感系统,尤其是自动驾驶中的协作感知系统,可以通过多视角传感功能显著提高跟踪精度并减少盲点。然而,其有限的信道容量和感知数据的冗余性带来了巨大挑战,影响了协作推理任务的性能。为了解决这些问题,我们为协作式边缘视频分析引入了优先级信息瓶颈(PIB)框架。我们首先提出了基于优先级的推理机制,该机制综合考虑了信噪比(SNR)和摄像机对感兴趣区域(RoI)的覆盖范围。为了实现高效推理,PIB 减少了空间和时间域的视频冗余,只传输下游推理任务所需的基本信息。这样就无需在边缘服务器上重建视频,同时保持低延迟。具体来说,它采用确定性信息瓶颈(IB)方法,在特征信息量和通信成本之间取得平衡,从而获得紧凑的任务相关特征。考虑到基于 IB 的高维数据目标所带来的计算挑战,我们采用了变分近似法进行可行性优化。与 TOCOM-TEM、JPEG 和 HEVC 相比,PIB 在平均物体检测准确率(MODA)方面实现了高达 15.1% 的改进,并且在边缘相机遭遇不良信道条件时,通信成本降低了 66.7%。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
PIB: Prioritized Information Bottleneck Framework for Collaborative Edge Video Analytics
Collaborative edge sensing systems, particularly in collaborative perception systems in autonomous driving, can significantly enhance tracking accuracy and reduce blind spots with multi-view sensing capabilities. However, their limited channel capacity and the redundancy in sensory data pose significant challenges, affecting the performance of collaborative inference tasks. To tackle these issues, we introduce a Prioritized Information Bottleneck (PIB) framework for collaborative edge video analytics. We first propose a priority-based inference mechanism that jointly considers the signal-to-noise ratio (SNR) and the camera's coverage area of the region of interest (RoI). To enable efficient inference, PIB reduces video redundancy in both spatial and temporal domains and transmits only the essential information for the downstream inference tasks. This eliminates the need to reconstruct videos on the edge server while maintaining low latency. Specifically, it derives compact, task-relevant features by employing the deterministic information bottleneck (IB) method, which strikes a balance between feature informativeness and communication costs. Given the computational challenges caused by IB-based objectives with high-dimensional data, we resort to variational approximations for feasible optimization. Compared to TOCOM-TEM, JPEG, and HEVC, PIB achieves an improvement of up to 15.1\% in mean object detection accuracy (MODA) and reduces communication costs by 66.7% when edge cameras experience poor channel conditions.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信