RI:Small: Learning shape features with deep neural networks
RI:Small:使用深度神经网络学习形状特征
基本信息
- 批准号:1814745
- 负责人:
- 金额:$ 45万
- 依托单位:
- 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:2018-09-01 至 2022-08-31
- 项目状态:已结题
- 来源:
- 关键词:
项目摘要
This project investigates how to effectively learn shape features with deep neural networks from images. It has been commonly believed that features learned by deep neural networks from images include texture, color, and shape of objects. Although visualizations of learned features demonstrate that contours of objects are extracted in the process of deep learning, our preliminary results provide clear arguments that 2D shape features are not well captured by current deep neural networks. This project develops a framework for effective learning of shape features with deep neural networks. The research brings new insights to a core problem in computer vision: shape understanding, which relates to many subfields in computer vision ranging from low-level tasks, such as segmentation and image statistics, to high-level ones, such as visual retrieval and object detection in images. The project includes plan to deploy the research results directly to applications such as biodiversity study (species recognition). The project also involves high school students and undergraduates in research.This project conducts both theoretical and experimental research to gain better understanding why shape features are not well captured by current deep neural networks. Then it develops new learning strategies specifically targeted for shape features by following two main alternatives: (1) constraining the filter learning for Convolutional Neural Networks so that they are more contour focused, and (2) designing special structures of Deep Neural Networks for learning shape representation. The project designs circular sequential networks for silhouette-based shape classification, which encode naturally contour context information while implicitly performing contour matching. It also extends these networks to sketches, which are composed of both closed and open contours. Attention models are investigated on shapes to analyze roles of parts in shape representations so as to improve further shape matching and recognition algorithms.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.
该项目研究如何使用深度神经网络从图像中有效地学习形状特征。人们普遍认为,深度神经网络从图像中学习到的特征包括物体的纹理、颜色和形状。虽然学习特征的可视化表明,物体的轮廓是在深度学习过程中提取的,但我们的初步结果提供了明确的论据,即当前的深度神经网络不能很好地捕捉2D形状特征。该项目开发了一个框架,用于使用深度神经网络有效学习形状特征。这项研究为计算机视觉的核心问题带来了新的见解:形状理解,它涉及计算机视觉中的许多子领域,从低级任务(如分割和图像统计)到高级任务(如视觉检索和图像中的对象检测)。该项目包括将研究成果直接应用于生物多样性研究(物种识别)等应用的计划。该项目还涉及高中生和本科生的研究。该项目进行理论和实验研究,以更好地理解为什么当前的深度神经网络不能很好地捕捉形状特征。然后,它通过以下两种主要选择开发了专门针对形状特征的新学习策略:(1)限制卷积神经网络的滤波器学习,使其更加关注轮廓,以及(2)设计深度神经网络的特殊结构来学习形状表示。该项目设计了基于轮廓的形状分类,自然编码轮廓上下文信息,同时隐式执行轮廓匹配的圆形顺序网络。它还将这些网络扩展到草图,这些草图由封闭和开放的轮廓组成。该奖项反映了NSF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(34)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)
GTCaR: Graph Transformer for Camera Re-localization
- DOI:10.1007/978-3-031-20080-9_14
- 发表时间:2022
- 期刊:
- 影响因子:0
- 作者:Xinyi Li;Haibin Ling
- 通讯作者:Xinyi Li;Haibin Ling
Osteoporosis Prescreening and Bone Mineral Density Prediction using Dental Panoramic Radiographs
使用牙科全景X光片进行骨质疏松症预筛查和骨矿物质密度预测
- DOI:10.1109/embc46164.2021.9630183
- 发表时间:2021
- 期刊:
- 影响因子:0
- 作者:Singh, Yasha;Atulkar, Vivek;Ren, Jiaxiang;Yang, Jie;Fan, Heng;Latecki, Longin Jan;Ling, Haibin
- 通讯作者:Ling, Haibin
FAMNet: Learning Feature, Affinity And Multi-dimensional Assignment For Online Multiple Object Tracking
FAMNet:在线多目标跟踪的学习特征、亲和力和多维分配
- DOI:
- 发表时间:2019
- 期刊:
- 影响因子:0
- 作者:Chu, Peng;Ling, Haibin
- 通讯作者:Ling, Haibin
Online Multi-Object Tracking with Instance-Aware Single-Object Tracking and Dynamic Model Refreshment
具有实例感知单对象跟踪和动态模型刷新的在线多对象跟踪
- DOI:
- 发表时间:2019
- 期刊:
- 影响因子:0
- 作者:Chu, Peng;Fan, Heng;Tan, Chiu C.;Ling, Haibin
- 通讯作者:Ling, Haibin
Clustered Object Detection in Aerial Images
- DOI:10.1109/iccv.2019.00840
- 发表时间:2019-04
- 期刊:
- 影响因子:0
- 作者:Fan Yang;Heng Fan;Peng Chu;Erik Blasch;Haibin Ling
- 通讯作者:Fan Yang;Heng Fan;Peng Chu;Erik Blasch;Haibin Ling
{{
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 }}
Longin Jan Latecki其他文献
Graph and Subspace Learning for Domain Adaptation
- DOI:
- 发表时间:
2015 - 期刊:
- 影响因子:0
- 作者:
Longin Jan Latecki - 通讯作者:
Longin Jan Latecki
UITI2007-University Information Technical Interchange Review Meeting
UITI2007-高校信息技术交流评审会
- DOI:
- 发表时间:
2008 - 期刊:
- 影响因子:0
- 作者:
Franques;Roger Williams;S. Schubert;J. Bloch;A. Ostrogorsky;A. Burger;Zhong He;J. Derby;Kelvin G. Lynn;J. D. Pruneda;D. McGregor;P. Lucas;K. Richardson;S. Hauck;K. Webb;M. Richardson;S. Sharpe;L. Carin;G. Wolberg;J. Gunther;T. Moon;Longin Jan Latecki;S. Balkır;I. Paschalidis;A. Garrett;G. Tepper;Z. Pizlo;G. Williams;J. Ryan;A. Maccabe;Jun Qi;M. Hoffman - 通讯作者:
M. Hoffman
Polygonal approximation of laser range data based on perceptual grouping and EM
基于感知分组和EM的激光测距数据的多边形逼近
- DOI:
- 发表时间:
2006 - 期刊:
- 影响因子:0
- 作者:
Longin Jan Latecki;Rolf Lakämper - 通讯作者:
Rolf Lakämper
Semi-Supervised Learning on an Augmented Graph with Class Labels
带有类标签的增强图的半监督学习
- DOI:
10.3233/978-1-61499-672-9-1571 - 发表时间:
2016 - 期刊:
- 影响因子:0
- 作者:
Nan Li;Longin Jan Latecki - 通讯作者:
Longin Jan Latecki
Using spatiotemporal blocks to reduce the uncertainty in detecting and tracking moving objects in video
使用时空块减少检测和跟踪视频中移动对象的不确定性
- DOI:
10.1504/ijista.2006.009914 - 发表时间:
2006 - 期刊:
- 影响因子:0
- 作者:
Longin Jan Latecki;V. Megalooikonomou;Roland Miezianko;D. Pokrajac - 通讯作者:
D. Pokrajac
Longin Jan Latecki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('Longin Jan Latecki', 18)}}的其他基金
RI: Medium: Collaborative Research: Object and Activity Recognition as the Maximum Weight Subgraph Problem with Mutual Exclusion Constraints
RI:中:协作研究:对象和活动识别作为具有互斥约束的最大权重子图问题
- 批准号:
1302164 - 财政年份:2013
- 资助金额:
$ 45万 - 项目类别:
Continuing Grant
EAGER: Solving Markov Random Fields with Mutual Exclusion Constraints
EAGER:求解具有互斥约束的马尔可夫随机场
- 批准号:
1257024 - 财政年份:2012
- 资助金额:
$ 45万 - 项目类别:
Standard Grant
CDI-Type II: Collaborative Research: Perception of Scene Layout by Machines and Visually Impaired Users
CDI-Type II:协作研究:机器和视障用户对场景布局的感知
- 批准号:
1027897 - 财政年份:2010
- 资助金额:
$ 45万 - 项目类别:
Standard Grant
Collaborative Research: Recovery of 3D Shapes from Single Views
合作研究:从单一视图恢复 3D 形状
- 批准号:
0924164 - 财政年份:2009
- 资助金额:
$ 45万 - 项目类别:
Continuing Grant
Collaborative Research: Simultaneous Contour Grouping and Medial Axis Estimation
协作研究:同时轮廓分组和中轴估计
- 批准号:
0812118 - 财政年份:2008
- 资助金额:
$ 45万 - 项目类别:
Standard Grant
Collaborative Research: From Edge Pixels to Recognition of Parts of Object Contours
协作研究:从边缘像素到物体轮廓部分的识别
- 批准号:
0534929 - 财政年份:2005
- 资助金额:
$ 45万 - 项目类别:
Continuing Grant
US-Germany Cooperative Research: Robot Localization and Robot Mapping Based on Shape Matching
美德合作研究:基于形状匹配的机器人定位与机器人建图
- 批准号:
0331786 - 财政年份:2003
- 资助金额:
$ 45万 - 项目类别:
Standard Grant
相似国自然基金
昼夜节律性small RNA在血斑形成时间推断中的法医学应用研究
- 批准号:
- 批准年份:2024
- 资助金额:0.0 万元
- 项目类别:省市级项目
tRNA-derived small RNA上调YBX1/CCL5通路参与硼替佐米诱导慢性疼痛的机制研究
- 批准号:n/a
- 批准年份:2022
- 资助金额:10.0 万元
- 项目类别:省市级项目
Small RNA调控I-F型CRISPR-Cas适应性免疫性的应答及分子机制
- 批准号:32000033
- 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
Small RNAs调控解淀粉芽胞杆菌FZB42生防功能的机制研究
- 批准号:31972324
- 批准年份:2019
- 资助金额:58.0 万元
- 项目类别:面上项目
变异链球菌small RNAs连接LuxS密度感应与生物膜形成的机制研究
- 批准号:81900988
- 批准年份:2019
- 资助金额:21.0 万元
- 项目类别:青年科学基金项目
基于small RNA 测序技术解析鸽分泌鸽乳的分子机制
- 批准号:31802058
- 批准年份:2018
- 资助金额:26.0 万元
- 项目类别:青年科学基金项目
肠道细菌关键small RNAs在克罗恩病发生发展中的功能和作用机制
- 批准号:31870821
- 批准年份:2018
- 资助金额:56.0 万元
- 项目类别:面上项目
Small RNA介导的DNA甲基化调控的水稻草矮病毒致病机制
- 批准号:31772128
- 批准年份:2017
- 资助金额:60.0 万元
- 项目类别:面上项目
基于small RNA-seq的针灸治疗桥本甲状腺炎的免疫调控机制研究
- 批准号:81704176
- 批准年份:2017
- 资助金额:20.0 万元
- 项目类别:青年科学基金项目
水稻OsSGS3与OsHEN1调控small RNAs合成及其对抗病性的调节
- 批准号:91640114
- 批准年份:2016
- 资助金额:85.0 万元
- 项目类别:重大研究计划
相似海外基金
Collaborative Research: RI: Small: Foundations of Few-Round Active Learning
协作研究:RI:小型:少轮主动学习的基础
- 批准号:
2313131 - 财政年份:2023
- 资助金额:
$ 45万 - 项目类别:
Standard Grant
Collaborative Research: RI: Small: Deep Constrained Learning for Power Systems
合作研究:RI:小型:电力系统的深度约束学习
- 批准号:
2345528 - 财政年份:2023
- 资助金额:
$ 45万 - 项目类别:
Standard Grant
Collaborative Research: RI: Small: End-to-end Learning of Fair and Explainable Schedules for Court Systems
合作研究:RI:小型:法院系统公平且可解释的时间表的端到端学习
- 批准号:
2232055 - 财政年份:2023
- 资助金额:
$ 45万 - 项目类别:
Standard Grant
CISE-ANR: RI: Small: Numerically efficient reinforcement learning for constrained systems with super-linear convergence (NERL)
CISE-ANR:RI:小:具有超线性收敛 (NERL) 的约束系统的数值高效强化学习
- 批准号:
2315396 - 财政年份:2023
- 资助金额:
$ 45万 - 项目类别:
Standard Grant
Collaborative Research: RI: Small: End-to-end Learning of Fair and Explainable Schedules for Court Systems
合作研究:RI:小型:法院系统公平且可解释的时间表的端到端学习
- 批准号:
2232054 - 财政年份:2023
- 资助金额:
$ 45万 - 项目类别:
Standard Grant
RI: Small: Approximate Inference for Planning and Reinforcement Learning
RI:小:规划和强化学习的近似推理
- 批准号:
2246261 - 财政年份:2023
- 资助金额:
$ 45万 - 项目类别:
Standard Grant
Collaborative Research: RI: Small: Foundations of Few-Round Active Learning
协作研究:RI:小型:少轮主动学习的基础
- 批准号:
2313130 - 财政年份:2023
- 资助金额:
$ 45万 - 项目类别:
Standard Grant
Collaborative Research: RI: Small: End-to-end Learning of Fair and Explainable Schedules for Court Systems
合作研究:RI:小型:法院系统公平且可解释的时间表的端到端学习
- 批准号:
2334936 - 财政年份:2023
- 资助金额:
$ 45万 - 项目类别:
Standard Grant
Collaborative Research: RI: AF: Small: Long-Term Impact of Fair Machine Learning under Strategic Individual Behavior
合作研究:RI:AF:小:战略性个人行为下公平机器学习的长期影响
- 批准号:
2202699 - 财政年份:2022
- 资助金额:
$ 45万 - 项目类别:
Standard Grant
Collaborative Research: RI: Small: Advancing Theory and Practice of Trustworthy Machine Learning via Bi-Level Optimization
合作研究:RI:小型:通过双层优化推进可信机器学习的理论和实践
- 批准号:
2207052 - 财政年份:2022
- 资助金额:
$ 45万 - 项目类别:
Standard Grant