CRII: OAC: Data Collection Infrastructure for Panoramic Video Monitoring in Wildlife Science

CRII：OAC：野生动物科学全景视频监控数据收集基础设施

• 批准号: 2151463
• 负责人: Zhisheng Yan
• 金额: $ 17.5万
• 依托单位: George Mason University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-10-01 至 2023-05-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2151463&HistoricalAwards=false
• 关键词: CRII; OAC; Data; Collection; Infrastructure

Wildlife monitoring has significant scientific and societal impacts. By utilizing remote cameras, biologists and ecologists can monitor and manage wildlife in order to prevent the transmission of zoonotic disease from animals and the invasion of wildlife on crops and livestock. However, current cyberinfrastructure (CI) in wildlife monitoring is limited to normal angle videos with a limited field of view and has caused missing the recording of important events that occurred outside of the direction being filmed. Moreover, existing remote cameras only allow the recording of short videos for a few minutes and thus cannot document many hours of wildlife activity in the monitoring zone. This project proposes methods for panoramic video monitoring that capture 360 degree uninterrupted videos to document complete wildlife activities. The project will allow wildlife scientists to access high fidelity monitoring data in both the spatial and temporal domains. Panoramic videos will not only capture comprehensive details on and near the monitoring site, but also depict the monitoring context of the data collection. The abundant research data and metadata embedded in panoramic videos will enhance the productivity of biologists and ecologists. If successful, the proposed wildlife monitoring CI will accelerate the adoption of panoramic data collection in other field research such as agriculture and archeology. The research outcomes, including the datasets generated and the software developed, will provide an interdisciplinary opportunity for undergraduate research, course curriculum development, and high school outreach activities, especially for underrepresented groups.This project investigates a video collection cyberinfrastructure to enable panoramic wildlife monitoring. The design objective is to archive days to weeks of high resolution video data for long lived monitoring under the limited storage and energy constraints of remote cameras. To this end, this project proposes a framework for collaborative local and networked storage. First, we propose camera computing strategies to understand the scientific value of monitoring content and maximally compress the video with negligible overhead. This would mitigate the overall need for storage. Second, we propose a networked storage scheme to address the intermittent nature of the network in the wild, where only partial video is transported while the remaining video is generated in the receiver. We then schedule compressed video tiles for local storage or networked storage by orchestrating the storage, network and battery resources. Finally, we will develop and deploy the panoramic video monitoring in real wildlife research. We will validate the CI on the Savannah River site and assist wildlife scientists to study the impacts of animal interaction on disease transmission.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

野生动物监测具有重要的科学和社会影响。通过使用远程摄像机，生物学家和生态学家可以监测和管理野生动物，以防止动物传播人畜共患疾病以及野生动物对作物和牲畜的入侵。然而，目前野生动物监测的网络基础设施（CI）仅限于正常角度的视频，视野有限，导致错过了拍摄方向之外发生的重要事件的记录。此外，现有的远程摄像机只能录制几分钟的短视频，因此无法记录监测区内野生动物的长时间活动。这个项目提出了全景视频监控的方法，捕捉360度不间断的视频，以记录完整的野生动物活动。该项目将允许野生动物科学家获得空间和时间领域的高保真监测数据。全景视频不仅可以捕捉监测现场和附近的全面细节，还可以描述数据收集的监测背景。丰富的研究数据和元数据嵌入全景视频将提高生物学家和生态学家的生产力。如果成功，提议的野生动物监测CI将加速在农业和考古学等其他领域研究中采用全景数据收集。研究成果，包括生成的数据集和开发的软件，将为本科研究、课程课程开发和高中外展活动提供一个跨学科的机会，特别是对代表性不足的群体。这个项目研究了一个视频收集网络基础设施，以实现全景野生动物监测。设计目标是在远程摄像机有限的存储和能量限制下，存档数天至数周的高分辨率视频数据，以便进行长期监控。为此，本项目提出了一个协作本地和网络存储的框架。首先，我们提出了摄像机计算策略，以了解监控内容的科学价值，并在可以忽略的开销下最大限度地压缩视频。这将减少对存储的总体需求。其次，我们提出了一种网络存储方案，以解决野外网络的间歇性，其中只有部分视频被传输，而剩余的视频在接收器中生成。然后，我们通过协调存储、网络和电池资源，为本地存储或网络存储安排压缩视频块。最后，我们将开发和部署全景视频监控在实际野生动物研究。我们将在萨凡纳河现场验证CI，并协助野生动物科学家研究动物相互作用对疾病传播的影响。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Deep Contextualized Compressive Offloading for Images

• DOI: 10.1145/3485730.3493452
• 发表时间: 2021-11
• 期刊: Proceedings of the 19th ACM Conference on Embedded Networked Sensor Systems
• 作者: Bo Chen;Zhisheng Yan;Hongpeng Guo;Zhe Yang;Ahmed Ali-Eldin;Prashant J. Shenoy;K. Nahrstedt

DAO: Dynamic Adaptive Offloading for Video Analytics

• DOI: 10.1145/3503161.3548249
• 发表时间: 2022-10
• 期刊: Proceedings of the 30th ACM International Conference on Multimedia
• 作者: Taslim Murad;Anh Nguyen;Zhisheng Yan

Context-aware image compression optimization for visual analytics offloading

用于视觉分析卸载的上下文感知图像压缩优化

• DOI: 10.1145/3524273.3528178
• 发表时间: 2022
• 期刊: MMSys '22: Proceedings of the 13th ACM Multimedia Systems Conference
• 作者: Chen, Bo;Yan, Zhisheng;Nahrstedt, Klara
• 通讯作者: Nahrstedt, Klara