位置：立项数据库 > 立项详情页
数据驱动的多维媒体感知与理解
  • 项目名称：数据驱动的多维媒体感知与理解
  • 项目类别：重点项目
  • 批准号：61035002
  • 申请代码：F030403
  • 项目来源：国家自然科学基金
  • 研究期限：2011-01-01-2014-12-31
  • 项目负责人：戴琼海
  • 负责人职称：教授
  • 依托单位：清华大学
  • 批准年度：2010
中文摘要：

现实场景的三维重建在军事、航天、医疗、影视制作等领域有着广泛的应用前景。利用稀疏分布的低分辨率相机感知高速运动、动态场景的光照迁移与逼真再现是三维重建的国际前沿挑战难题，通过现有采集机制与视觉处理方法难以刻画"大规模"、"变光照"、"高动态"的多维媒体特性。涉及到的核心问题包括高动态复杂对象空时光采样，多维媒体几何及光照特性的感知与理解、重生成与优化等。本项目瞄准多维媒体数据的几何、光照与动态特性，探索大规模数据采样的机制，通过时间解耦、压缩采样、自动补偿高精密采样实现场景的空时光数据捕获，采用数据驱动的方法，结合物理特性建模与图像数据分析，解决场景空时光感知与理解的关键问题，同时研究多维媒体的视觉质量评价方法，对场景的高效采样、逼真重建进行反馈指导。力图揭示多维媒体特性的表示机理，建立数据驱动的多维媒体感知与理解平台。在理论和关键技术研究上取得突破，培养一支多维媒体领域的研究团队。

中文主题词： 三维重建；数据驱动；多维媒体；视觉处理；
英文摘要：

3D reconstruction；data driven；multidimensional media；vision processing；

英文主题词： 3D reconstruction；data driven；multidimensional media；vision processing；
结论摘要：

现实场景的三维重建在军事、航天、医疗、影视制作等领域有着广泛的应用前景。利用稀疏分布的低分辨率相机感知高速运动、动态场景的光照迁移与逼真再现是三维重建的国际前沿挑战难题，通过现有采集机制与视觉处理方法难以刻画"大规模"、"变光照"、"高动态"的多维媒体特性。涉及到的核心问题包括高动态复杂对象空时光采样，多维媒体几何及光照特性的感知与理解、重生成与优化等。本项目瞄准多维媒体数据的几何、光照与动态特性，探索大规模数据采样的机制，通过时间解耦、压缩采样、自动补偿高精密采样实现场景的空时光数据捕获，采用数据驱动的方法，结合物理特性建模与图像数据分析，解决场景空时光感知与理解的关键问题，同时研究多维媒体的视觉质量评价方法，对场景的高效采样、逼真重建进行反馈指导。力图揭示多维媒体特性的表示机理，建立数据驱动的多维媒体感知与理解平台。在理论和关键技术研究上取得突破，培养一支多维媒体领域的研究团队。

成果综合统计
成果类型
数量
  • 期刊论文
  • 会议论文
  • 专利
  • 获奖
  • 著作
  • 57
  • 24
  • 0
  • 0
  • 0
期刊论文
Bilateral blue noise sampling
Adaptive Compressed Sensing Recovery Utilizing the Property of Signal's Autocorrelations
基于视频的现实人物运动创作
Camera Constraint-Free View-Based 3-D Object Retrieval
Solving Under-constrained Assembly Problems Incrementally Using a Kinematic Method
Coded aperture pair for quantitative phase imaging
Collaborative color calibration for multi-camera systems
曲面二阶几何连续性的混合曲率评价与可视化
Performance Capture of High-Speed Motion Using Staggered Multi-View Recording
3-D Object Retrieval and Recognition With Hypergraph Analysis
Converting 2D Video to 3D: An Efficient Path to a 3D Experience
Less is More: Efficient 3-D Object Retrieval With Query View Selection
Content-adaptive high-resolution hyperspectral video acquisition with a hybrid camera system
Dual-coded compressive hyperspectral imaging
High-dimensional camera shake removal with given depth map
Registration of point clouds using sample-sphere and adaptive distance restriction
Polyline approach for approximating Hausdorff distance between planar free-form curves
Complexity Reduction and Performance Improvement for Geometry Partitioning in Video Coding
Low-Rank Structure Learning via Nonconvex Heuristic Recovery
An Overview of Computational Sparse Models and Their Applications in Artificial Intelligence
Acquisition of High Spatial and Spectral Resolution Video with a Hybrid Camera System
Joint Non-Gaussian Denoising and Superresolving of Raw High Frame Rate Videos
Reweighted low-rank matrix recovery and its application in image restoration
Separable coded aperture for depth from a single image
A parametric model for describing the correlation between single color images and depth maps
Video-based characters: creating new human performances from a multi-view video database
3D model retrieval using weighted bipartite graph matching. Signal Processing: Image Communication
Epsilon-G2 B-spline surface interpolation
Efficient depth-of-field rendering with adaptive sampling and multi-scale reconstruction
An improved example-driven symbol recognition approach in engineering drawings
Semi-Automatic 2D-to-3D Conversion Using Disparity Propagation
Calculating Jacobian coefficients of primitive constraints with respect to Euler parameters
Continuity transition with a single regular curved-knot spline surface
Projecting points onto planar parametric curves by local biarc approximation
A linear algorithm for pseudo recognition in 3D reconstruction
Converting sectional views to three orthographic views to reconstruct 3D models
Ultra-fast Lensless Computational Imaging through 5D Frequency Analysis of Time-resolved Light Trans
A new method for identifying and validating features from 2D sectional views
Algorithm for Directional Projection of Parametric Curves onto B-spline Surfaces
从压缩传感到低秩矩阵恢复:理论与应用
基于2-邻域局部结构的矢量图符号模糊识别方法
Noisy Depth Maps Fusion for Multiview Stereo Via Matrix Completion
A Data-Driven Approach for Facial Expression Retargeting in Video
From Compressed Sensing to Low-rank Matrix Recovery: Theory and Applications
Video-object segmentation and 3D-trajectory estimation for monocular video sequences
Polar NURBS surface with curvature continuity
会议论文
Transparent object reconstruction via coded transport of intensity
A Single Frame Super-Resolution Method Based on Matrix Completion
3D SPATIAL RECONSTRUCTION AND COMMUNICATION FROM VISION FIELD
Hybrid image deblurring by fusing edge and power spectrum information
Indoor structure understanding from single 360 cylindrical panoramic image
A NOVEL 2D-TO-3D SCHEME BY VISUAL ATTENTION AND OCCLUSION ANALYSIS
Dynamic shape capture via periodical-illumination optical flow estimation and multi-view photometric
Adaptive information interactive mechanism for multi-UAV visual navigation
Iterative feedback estimation of depth and radiance from defocused images
Frequency analysis of transient light transport with applications in bare sensor imaging
Implicit surface reconstruction based on adaptive clustering
Multi-resolution mesh fitting by B-spline surfaces for reverse engineering
An example-driven symbol recognition approach based on key features in engineering drawings
Visual words assignment on a graph via minimal mutual information loss
Multi-view photometric stereo of non-Lambertian surface under general illuminations
Vision field capture for advanced 3DTV applications
High Resolution Multispectral Video Capture with a Hybrid Camera System
GRAVITY-ORIENTED NONLINEAR STEREO RECTIFICATION FOR STEREOSCOPIC 3D
Visual information, sparse decomposition, and transmission for multi-UAV visual navigation
ROBUST FOCAL LENGTH ESTIMATION BASED ON MINIMAL SOLUTION METHOD
A NOVEL METHOD FOR 2D-TO-3D VIDEO CONVERSION USING BI-DIRECTIONAL MOTION ESTIMATION
Fourier analysis on transient imaging with a multifrequency time-of-flight camera
Recovering scene geometry under wavy fluid via distortion and defocus analysis
Exploring aligned complementary image pair for blind motion deblurring
相关项目
复杂过程系统的递阶神经网络结构研究
期刊论文 26 会议论文 2 专利 2
真实场景与虚拟物体混合的增强视频技术研究
期刊论文 22 会议论文 1
复杂生产过程智能调度与优化
期刊论文 50 会议论文 14
基于单CCD相机自由拍摄的三维柔性测量关键技术研究
期刊论文 28 会议论文 2 获奖 2
复杂生产过程基于数据的优化调度理论与方法及其在冶金工业中的应用研究
期刊论文 62 会议论文 19
AVR理论及实景虚化技术
期刊论文 3 会议论文 11
基于形状文法和多源数据融合的三维建筑高效重构方法研究
期刊论文 4 会议论文 6 专利 4 著作 1
基于视觉的织物疵点三维检测和三维识别原理研究
期刊论文 19 会议论文 5 获奖 4
作物根系三维形态原位测量方法的研究
期刊论文 11 会议论文 4
戴琼海的项目
移动普适多媒体国际会议
数字媒体网络系统及关键技术的研究
期刊论文 26 会议论文 14 获奖 2
信源编码与信道编码
期刊论文 30 会议论文 34
2007中美关系研讨会
多维多尺度高分辨率计算摄像仪器
期刊论文 10
第八届国际广播电视技术讨论会
图象编码会议
FMmlet变换的空间对称性、多维变换、快速算法及应用
　复杂条件下飞行器进近可视导航的基础理论研究
期刊论文 89
计算摄像学中的新理论及新方法
期刊论文 1