Photogrammetric Computer Vision and Image Analysis

Terms of Reference

• Algorithms for geometric analysis of image data regardless of scale;
• Geometric analyses of IR, SAR, IFSAR and LIDAR;
• Automated feature and attribute extraction techniques and methodologies from multi-sensor, multi-resolution, multi-spectral, hyperspectral, and multi-temporal imagery;
• Fundamental research into image understanding for object detection, recognition, identification and reconstruction;
• DEM generation and integration of three-dimensional modeling concepts into image analysis processes;
• Integration of spatial information systems and object models for object recognition;
• Sensor pose determination (including auxiliary information);
• Projective and multi-view geometry;
• Image sequence analysis;
• Algorithms for including features in the orientation processes;
• Spatial, spectral and temporal properties of natural and human-formed objects.

WORKING GROUPS of ISPRS COMMISSION III

• WG III/1 - Orientation and Surface Reconstruction
• WG III/2 - 3D Point Cloud Processing
• WG III/3 - Image sequence analysis
• WG III/4 - 3D Scene Analysis
• WG III/5 - Computer Graphics and Remote Sensing

INTERCOMMISSION WGs of ISPRS COMMISSION III

• ICWG III/I - Sensor Modeling for Integrated Orientation and Navigation
• ICWG III/VII - Pattern Analysis in Remote Sensing

WG III/1 - Orientation and Surface Reconstruction

WG III/1  Terms of Reference

• Camera pose estimation
• Calibration and orientation without artificial targets
• Multi-view matching and surface reconstruction
• Evaluation of performance, reliability and generality of methods
• Robustness of orientation and matching

WG III/2 - 3D Point Cloud Processing

WG III/2  Terms of Reference

• Scene understanding from aerial, mobile, terrestrial and indoor point cloud data
• Change detection and map updating using multi-temporal point clouds
• Information extraction from point clouds, including segmentation and classification
• Registration of point clouds
• Machine learning algorithms in point cloud processing
• High performance computing for massive point data processing and analysis
• Analysis of full-waveform lidar data

WG III/3 - Image sequence analysis

WG III/3  Terms of Reference

• Camera and camera network calibration from image sequences including cameras with non-standard geometry and variable frame rates
• Models and methods to determine ego-motion, for navigation, geo-referencing and object reconstruction
• Detection, reconstruction, classification and tracking of single and multiple objects in image sequences
• Event reconstruction and analysis from image sequences, and from single and multiple video streams
• Quality assessment techniques for calibration, orientation and object detection from image sequences, including but not limited to time-series analysis at different epochs and resolutions
• Benchmarking of calibration, orientation and object detection with image sequences
• Change detection in time-series of images or 3D point clouds, including the analysis of landscape or object evolution of huge heterogeneous data sets

WG III/4 - 3D Scene Analysis

WG III/4  Terms of Reference

• Models and techniques for extracting objects (buildings, roads, cars, vegetation etc.) from aerial, spaceborne and terrestrial image, laser, and (In)SAR data
• Generation and update of high-resolution 3D city models and road databases
• Interpretation of terrestrial, aerial and spaceborne sensor data, possibly together with information from traditional cartographic products, CAD models, and urban GIS.
• Tools and models for integrating information about multiple object classes and their relations within complex scenes.
• Automatic and semi-automatic generation of urban models with level-of-detail (LOD) and attributes
• Analysis of the trade-off between geometry and radiometry / texture for visualization
• Assessment of efficiency and quality, and of their dependence on the quality of the input data

WG III/5 - Computer Graphics and Remote Sensing

WG III/5  Terms of Reference

Rendering / Visualization:

• Image-based rendering : geolocalized and calibrated images as plenoptic function samples, with or without possibly uncertain 3D cues (3D-models, 3D point clouds...).
• Point-based rendering : Datastructures and processing
• Algorithms for Augmented Reality and Interactive Environments
• Algorithms for Web Visualization (e.g. HTML5 Canvas, WebGL)
• Out-of-Core rendering large 3D-scenes, large models, and point clouds
• Physical simulation of the radiative transfer involved in image or Lidar acquisition (relighting, generation of synthetic datasets for analysis)
• Spatial data structures and geometric algorithms

Inverse Rendering / Analysis :

• Reflectance estimation for physical or non-physical image relighting (eg: shadow removal)
• Reflectance estimation as a material feature for subsequent analysis (eg: classification)
• Use of graphics hardware (GPU) for solving computationally expensive problems

ICWG III/I - Sensor Modeling for Integrated Orientation and Navigation

ICWG III/I  Terms of Reference

• General rigorous modeling of positioning and attitude sensors.
• Modeling of MEMS sensors for photogrammetric orientation.
• Stochastic Differential Equations in orientation and calibration for photogrammetric applications.
• Spatial and temporal modeling of non-standard sensors (low-cost, new-geometry, combined-geometry configurations).
• Block adjustment-oriented modeling of hybrid orientation and calibration systems
• Photogrammetric/LiDAR orientation and calibration: measurement techniques and models.

ICWG III/VII - Pattern Analysis in Remote Sensing

ICWG III/VII  Terms of Reference

• Automatic identification and learning of 2D and 3D patterns in uni-modal and multi-modal remote sensing data, e.g. multi-scale aerial and satellite data; multi- and hyperspectral data; SAR-, radargrammetric and SAR-tomography data
• Automatic identification and learning of temporal patterns in remote sensing data, e.g. image-based flow estimation and learning from InSAR data (traffic, glaciers, currents, etc.); analysis-by-synthesis approaches for motion and deformation modeling with passive and active sensors
• Integration of radiometry and radiometric models into pattern recognition; radiometrically enhanced object models for range-intensity images and sequences; integration of SAR-simulation into SAR-image and analysis