Deep Domain Adaptation and Fusion for Person Recognition in the Wild
用于野外人员识别的深度域适应和融合
基本信息
- 批准号:543663-2019
- 负责人:
- 金额:$ 5.1万
- 依托单位:
- 依托单位国家:加拿大
- 项目类别:Collaborative Research and Development Grants
- 财政年份:2020
- 资助国家:加拿大
- 起止时间:2020-01-01 至 2021-12-31
- 项目状态:已结题
- 来源:
- 关键词:
项目摘要
Nuvoola Inc. is a Canadian company that provides cloud managed services and solutions based on artificial intelligence. It seeks to develop software for tracking the location of individuals appearing in real-world video surveillance environments. In particular, Nuvoola seeks to develop accurate person re-identification (Re-ID) systems that allow recognizing a same person of interest appearing over time, and across multiple distributed cameras. Re-ID is a challenging problem because of variations in capture conditions, e.g., cameras viewpoint, pose, illumination, scale, motion blur, background, occlusion. State-of-the-art systems for Re-ID employ deep Siamese networks for matching, yet these deep models require supervised training on a large and diverse amount of labeled data to achieve a high level of accuracy and robustness. For real-world applications, however, Re-ID accuracy can decline if there is a considerable divergence between the capture conditions of development and operational environments. The first objective of this project is to investigate and develop deep learning architectures that are suitable for accurate domain adaptation (DA) of deep Siamese networks in real-world applications by learning domain-invariant representations. Some techniques in literature are relevant for this project but provide limited versatility and improvements on real-world Re-ID videos, where domain shifts are significant. The second objective is to investigate and develop specialized deep learning models for information fusion, and in particular on multi-modal, spatiotemporal, and context-based fusion. This project involves a cross-disciplinary team from ETS and Nuvoola Inc., and will allow to intensify the exchange of ideas and resources, and establish long-term collaborative links. Focusing on the design of deep learning models for visual recognition applications from large and mostly unlabeled datasets, this project will focus on state-of-the-art research. Significant findings of this research project will be disseminated in high caliber scientific journals and conferences and exploited in Nuvoola video analytics applications. It also allows training of highly qualified personnel to face current and future challenges in areas of strategic interest.
Nuvoola Inc.是一家加拿大公司,提供基于人工智能的云托管服务和解决方案。它寻求开发用于跟踪出现在现实世界视频监控环境中的个人位置的软件。特别是,Nuvoola寻求开发准确的个人重新识别(Re-ID)系统,该系统允许识别随着时间推移出现的同一个感兴趣的人,并跨多个分布式摄像头。由于拍摄条件的变化,例如摄像机的视点、姿势、光照、比例、运动模糊、背景、遮挡,重新ID是一个具有挑战性的问题。最先进的Re-ID系统使用深度暹罗网络进行匹配,但这些深度模型需要对大量不同的标签数据进行监督训练,以实现高水平的准确性和稳健性。然而,对于现实世界的应用程序,如果开发和操作环境的捕获条件之间存在相当大的差异,重新ID准确性可能会下降。该项目的第一个目标是通过学习领域不变的表示来研究和开发适合于实际应用中的深连体网络的精确域适应(DA)的深度学习体系结构。文献中的一些技术与该项目相关,但在真实世界的Re-ID视频上提供的通用性和改进有限,其中域转移是显著的。第二个目标是研究和开发专门的深度学习模型,用于信息融合,特别是多模式、时空和基于上下文的融合。该项目涉及ETS和Nuvoola Inc.的跨学科团队,将加强思想和资源的交流,并建立长期的合作联系。这个项目的重点是从大数据集和大多数未标记的数据集设计用于视觉识别应用的深度学习模型,该项目将专注于最先进的研究。这一研究项目的重大成果将在高质量的科学期刊和会议上传播,并在Nuvoola视频分析应用程序中加以利用。它还允许培训高素质的人员,以应对战略利益领域当前和未来的挑战。
项目成果
期刊论文数量(0)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)
数据更新时间:{{ journalArticles.updateTime }}
{{
item.title }}
{{ item.translation_title }}
- DOI:
{{ item.doi }} - 发表时间:
{{ item.publish_year }} - 期刊:
- 影响因子:{{ item.factor }}
- 作者:
{{ item.authors }} - 通讯作者:
{{ item.author }}
数据更新时间:{{ journalArticles.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ monograph.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ sciAawards.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ conferencePapers.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ patent.updateTime }}
Granger, Eric其他文献
Partially-supervised learning from facial trajectories for face recognition in video surveillance
- DOI:
10.1016/j.inffus.2014.05.006 - 发表时间:
2015-07-01 - 期刊:
- 影响因子:18.6
- 作者:
De-la-Torre, Miguel;Granger, Eric;Gorodnichy, Dmitry O. - 通讯作者:
Gorodnichy, Dmitry O.
Graphical EM for on-line learning of grammatical probabilities in radar Electronic Support
- DOI:
10.1016/j.asoc.2012.02.022 - 发表时间:
2012-08-01 - 期刊:
- 影响因子:8.7
- 作者:
Latombe, Guillaume;Granger, Eric;Dilkes, Fred A. - 通讯作者:
Dilkes, Fred A.
On the memory complexity of the forward-backward algorithm
- DOI:
10.1016/j.patrec.2009.09.023 - 发表时间:
2010-01-15 - 期刊:
- 影响因子:5.1
- 作者:
Khreich, Wael;Granger, Eric;Sabourin, Robert - 通讯作者:
Sabourin, Robert
A paired sparse representation model for robust face recognition from a single sample
- DOI:
10.1016/j.patcog.2019.107129 - 发表时间:
2020-04-01 - 期刊:
- 影响因子:8
- 作者:
Mokhayeri, Fania;Granger, Eric - 通讯作者:
Granger, Eric
Bag-Level Aggregation for Multiple-Instance Active Learning in Instance Classification Problems
- DOI:
10.1109/tnnls.2018.2869164 - 发表时间:
2019-05-01 - 期刊:
- 影响因子:10.4
- 作者:
Carbonneau, Marc-Andre;Granger, Eric;Gagnon, Ghyslain - 通讯作者:
Gagnon, Ghyslain
Granger, Eric的其他文献
{{
item.title }}
{{ item.translation_title }}
- DOI:
{{ item.doi }} - 发表时间:
{{ item.publish_year }} - 期刊:
- 影响因子:{{ item.factor }}
- 作者:
{{ item.authors }} - 通讯作者:
{{ item.author }}
{{ truncateString('Granger, Eric', 18)}}的其他基金
Deep Weakly-Supervised Neural Networks for Cross-Domain Video Recognition and Localization
用于跨域视频识别和定位的深度弱监督神经网络
- 批准号:
DGDND-2022-05397 - 财政年份:2022
- 资助金额:
$ 5.1万 - 项目类别:
DND/NSERC Discovery Grant Supplement
Deep Weakly-Supervised Neural Networks for Cross-Domain Video Recognition and Localization
用于跨域视频识别和定位的深度弱监督神经网络
- 批准号:
RGPIN-2022-05397 - 财政年份:2022
- 资助金额:
$ 5.1万 - 项目类别:
Discovery Grants Program - Individual
Deep Domain Adaptation and Fusion for Person Recognition in the Wild
用于野外人员识别的深度域适应和融合
- 批准号:
543663-2019 - 财政年份:2021
- 资助金额:
$ 5.1万 - 项目类别:
Collaborative Research and Development Grants
Adaptive Context-Based Systems for Face Recognition in Video Surveillance
视频监控中基于上下文的自适应人脸识别系统
- 批准号:
RGPIN-2016-06783 - 财政年份:2021
- 资助金额:
$ 5.1万 - 项目类别:
Discovery Grants Program - Individual
Adaptive Context-Based Systems for Face Recognition in Video Surveillance
视频监控中基于上下文的自适应人脸识别系统
- 批准号:
RGPIN-2016-06783 - 财政年份:2020
- 资助金额:
$ 5.1万 - 项目类别:
Discovery Grants Program - Individual
Detection of COVID-19 in Intelligent Building Occupancy Management
智能建筑占用管理中的 COVID-19 检测
- 批准号:
555212-2020 - 财政年份:2020
- 资助金额:
$ 5.1万 - 项目类别:
Alliance Grants
Deep Domain Adaptation and Fusion for Person Recognition in the Wild
用于野外人员识别的深度域适应和融合
- 批准号:
543663-2019 - 财政年份:2019
- 资助金额:
$ 5.1万 - 项目类别:
Collaborative Research and Development Grants
Adaptive Context-Based Systems for Face Recognition in Video Surveillance
视频监控中基于上下文的自适应人脸识别系统
- 批准号:
RGPIN-2016-06783 - 财政年份:2019
- 资助金额:
$ 5.1万 - 项目类别:
Discovery Grants Program - Individual
Adaptive Context-Based Systems for Face Recognition in Video Surveillance
视频监控中基于上下文的自适应人脸识别系统
- 批准号:
RGPIN-2016-06783 - 财政年份:2018
- 资助金额:
$ 5.1万 - 项目类别:
Discovery Grants Program - Individual
A Comparison of Siamese Convolutional Neural Networks for Person Re-Identification in Video Surveillance**
视频监控中人员重新识别的连体卷积神经网络的比较**
- 批准号:
533701-2018 - 财政年份:2018
- 资助金额:
$ 5.1万 - 项目类别:
Engage Grants Program
相似国自然基金
拟连续domain范畴的若干问题研究
- 批准号:12301583
- 批准年份:2023
- 资助金额:30 万元
- 项目类别:青年科学基金项目
格值蕴涵算子与Domain理论中的若干问题
- 批准号:12331016
- 批准年份:2023
- 资助金额:193.00 万元
- 项目类别:重点项目
Domain理论中概率幂构造的若干问题研究
- 批准号:12371457
- 批准年份:2023
- 资助金额:43.5 万元
- 项目类别:面上项目
To空间上Domain理论中若干问题研究
- 批准号:12261040
- 批准年份:2022
- 资助金额:28 万元
- 项目类别:地区科学基金项目
面向Jung-Tix问题的Domain理论与量化序理论研究
- 批准号:12231007
- 批准年份:2022
- 资助金额:235 万元
- 项目类别:重点项目
C2 DOMAIN PROTEIN 1 (C2DP1)基因家族在植物开花调控中的功能研究
- 批准号:n/a
- 批准年份:2022
- 资助金额:0.0 万元
- 项目类别:省市级项目
第四届Domain理论与拓扑学青年学者论坛
- 批准号:12242110
- 批准年份:2022
- 资助金额:5.00 万元
- 项目类别:专项项目
Domain理论中的收敛性与K-有界Sober空间范畴
- 批准号:
- 批准年份:2021
- 资助金额:30 万元
- 项目类别:青年科学基金项目
广义Domain结构的表示理论研究
- 批准号:12171149
- 批准年份:2021
- 资助金额:51 万元
- 项目类别:面上项目
拓扑空间的概率幂domain及相关问题研究
- 批准号:12001385
- 批准年份:2020
- 资助金额:24.0 万元
- 项目类别:青年科学基金项目
相似海外基金
Multilingual corpus construction and domain adaptation for low-resource machine translation
低资源机器翻译的多语言语料库构建和领域适应
- 批准号:
22KJ1724 - 财政年份:2023
- 资助金额:
$ 5.1万 - 项目类别:
Grant-in-Aid for JSPS Fellows
Revisiting Domain Adaptation in Video, within and Beyond Recognition
重新审视视频中的领域适应,在识别之内和之外
- 批准号:
2894961 - 财政年份:2023
- 资助金额:
$ 5.1万 - 项目类别:
Studentship
Domain adaptation approaches to unify established and emerging sequencing technologies
领域适应方法统一现有和新兴的测序技术
- 批准号:
10643544 - 财政年份:2023
- 资助金额:
$ 5.1万 - 项目类别:
Exploring unsupervised domain adaptation methods for automated linear disturbance mapping
探索自动线性扰动映射的无监督域适应方法
- 批准号:
577643-2022 - 财政年份:2022
- 资助金额:
$ 5.1万 - 项目类别:
Alliance Grants
Domain Adaptation Frameworks for Road Scene Segmentation in Unseen Environments
未知环境中道路场景分割的领域适应框架
- 批准号:
22K17976 - 财政年份:2022
- 资助金额:
$ 5.1万 - 项目类别:
Grant-in-Aid for Early-Career Scientists
A Mobile Game for Domain Adaptation and Deep Learning in Autism Healthcare
用于自闭症医疗领域适应和深度学习的手机游戏
- 批准号:
10596139 - 财政年份:2021
- 资助金额:
$ 5.1万 - 项目类别:
Domain adaptation for acoustic inspection of concrete structures
混凝土结构声学检测的域适应
- 批准号:
21K17829 - 财政年份:2021
- 资助金额:
$ 5.1万 - 项目类别:
Grant-in-Aid for Early-Career Scientists
A Mobile Game for Domain Adaptation and Deep Learning in Autism Healthcare
用于自闭症医疗领域适应和深度学习的手机游戏
- 批准号:
10443542 - 财政年份:2021
- 资助金额:
$ 5.1万 - 项目类别:
Domain adaptation and long-term maintenance of invisible maps to enable low-cost autonomous navigation of robots in urban environments
隐形地图的领域适应和长期维护,使机器人在城市环境中实现低成本自主导航
- 批准号:
21K14121 - 财政年份:2021
- 资助金额:
$ 5.1万 - 项目类别:
Grant-in-Aid for Early-Career Scientists
Research on Unsupervised Domain Adaptation to Reduce Labelling Costs
无监督领域适应降低标签成本的研究
- 批准号:
21K17799 - 财政年份:2021
- 资助金额:
$ 5.1万 - 项目类别:
Grant-in-Aid for Early-Career Scientists