权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

CIF: Small: Occlusion-Based Computational Imaging and Scene Analysis: Theory, Methods and Applications

CIF：小型：基于遮挡的计算成像和场景分析：理论、方法和应用

基本信息

批准号：
1816209
负责人：
Gregory Wornell
金额：
$ 49.88万
依托单位：
Massachusetts Institute of Technology
依托单位国家：
美国
项目类别：
Standard Grant
财政年份：
2018
资助国家：
美国
起止时间：
2018-10-01 至 2023-09-30
项目状态：
已结题

来源：
https://www.nsf.gov/awardsearch/showAward?AWD_ID=1816209&HistoricalAwards=false
关键词：
CIF Small Occlusion Based Computational

项目摘要

Humans have long sought ways to extend their vision, hearing, and other sensory capabilities, and there has been a long history of scientific and technological progress. These efforts have involved new physical devices that enhance existing human sensory capabilities (e.g., microscopes and stroboscopes), and and also the use of information from modalities humans cannot sense (e.g., radio frequency wave-forms and X-rays). Today we have the ability to jointly manipulate these sources of information both physically and computationally, via signal processing, to ever increasing degrees, and the resulting technologies are finding application in diverse domains. While interference effects would appear to limit the information that can be extracted, this need not be the case, as the familiar example of a pinhole camera illustrates. Motivated by such examples, this project develops, in a unified way, the role that such phenomena have to play in enhancing the amount of information that can be recovered from environments of interest. Potential applications include medicine, public safety, building security systems, autonomous vehicle safety systems, microscopy and astronomy, and virtual and augmented reality systems. The proposed research is a highly education-centered activity with a focus around the intellectual and professional development of students.The research investigates several instances and aspects of occlusion-based imaging problems, including: the analysis and optimization of interference phenomena; the development of optimum information extraction and inference of environmental parameters of interest; and investigation of blind methods for poorly characterized settings. The research emphasizes a common framework of analysis, and exploits a rich collection of tools from signal processing, statistical inference, systems theory, and information theory.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.

长期以来，人类一直在寻求扩展视觉、听觉和其他感官能力的方法，科学技术进步的历史也很悠久。这些努力涉及新的物理设备，以增强现有的人类感官能力（例如，显微镜和频闪仪），以及使用来自人类无法感知的模式的信息（例如，射频波形和x射线）。今天，我们有能力通过信号处理，在物理上和计算上共同操纵这些信息源，其程度不断提高，由此产生的技术正在各个领域得到应用。虽然干扰效应似乎会限制可以提取的信息，但事实并非如此，正如我们所熟悉的针孔相机的例子所示。在这些例子的激励下，本项目以统一的方式发展了这些现象在增加可从感兴趣的环境中恢复的信息量方面所起的作用。潜在的应用包括医学、公共安全、建筑安全系统、自动驾驶汽车安全系统、显微镜和天文学，以及虚拟和增强现实系统。拟议的研究是一项高度以教育为中心的活动，重点是学生的智力和专业发展。研究了基于遮挡的成像问题的几个实例和方面，包括：干扰现象的分析和优化；环境参数优化信息提取与推理技术的发展以及对特征不明显的环境进行盲法研究。该研究强调一个共同的分析框架，并利用了信号处理、统计推断、系统理论和信息论等丰富的工具集合。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

期刊论文数量（11）

专著数量（0）

科研奖励数量（0）

会议论文数量（0）

专利数量（0）

On Learning Continuous Pairwise Markov Random Fields

DOI：
发表时间：
2020-10
期刊：
ArXiv
影响因子：
0
作者：
Abhin Shah;Devavrat Shah;G. Wornell
通讯作者：
Abhin Shah;Devavrat Shah;G. Wornell

Learning New Tricks From Old Dogs: Multi-Source Transfer Learning From Pre-Trained Networks

DOI：
发表时间：
2019
期刊：
影响因子：
0
作者：
Joshua K. Lee;P. Sattigeri;G. Wornell
通讯作者：
Joshua K. Lee;P. Sattigeri;G. Wornell

ShadowCam: Real-Time Detection of Moving Obstacles Behind A Corner For Autonomous Vehicles

DOI：
10.1109/itsc.2018.8569569
发表时间：
2018-11
期刊：
2018 21st International Conference on Intelligent Transportation Systems (ITSC)
影响因子：
0
作者：
Felix Naser;Igor Gilitschenski;G. Rosman;Alexander Amini;F. Durand;A. Torralba;G. Wornell;W. Freeman;S. Karaman;D. Rus
通讯作者：
Felix Naser;Igor Gilitschenski;G. Rosman;Alexander Amini;F. Durand;A. Torralba;G. Wornell;W. Freeman;S. Karaman;D. Rus

What You Can Learn by Staring at a Blank Wall

DOI：
10.1109/iccv48922.2021.00233
发表时间：
2021-08
期刊：
2021 IEEE/CVF International Conference on Computer Vision (ICCV)
影响因子：
0
作者：
Prafull Sharma;M. Aittala;Y. Schechner;A. Torralba;G. Wornell;W. Freeman;F. Durand
通讯作者：
Prafull Sharma;M. Aittala;Y. Schechner;A. Torralba;G. Wornell;W. Freeman;F. Durand

Using Unknown Occluders to Recover Hidden Scenes

使用未知遮挡物恢复隐藏场景