Contributions to 3D Data Registration and Representation

Nowadays, new computers generation provides a high performance that enables to build computationally expensive computer vision applications applied to mobile robotics. Building a map of the environment is a common task of a robot and is an essential part to allow the robots to move through these environments. Traditionally, mobile robots used a combination of several sensors from different technologies. Lasers, sonars and contact sensors have been typically used in any mobile robotic architecture, however color cameras are an important sensor due to we want the robots to use the same information that humans to sense and move through the different environments. Color cameras are cheap and flexible but a lot of work need to be done to give robots enough visual understanding of the scenes. Computer vision algorithms are computational complex problems but nowadays robots have access to different and powerful architectures that can be used for mobile robotics purposes. The advent of low-cost RGB-D sensors like Microsoft Kinect which provide 3D colored point clouds at high frame rates made the computer vision even more relevant in the mobile robotics field. The combination of visual and 3D data allows the systems to use both computer vision and 3D processing and therefore to be aware of more details of the surrounding environment. The research described in this thesis was motivated by the need of scene mapping. Being aware of the surrounding environment is a key feature in many mobile robotics applications from simple robotic navigation to complex surveillance applications. In addition, the acquisition of a 3D model of the scenes is useful in many areas as video games scene modeling where well-known places are reconstructed and added to game systems or advertising where once you get the 3D model of one room the system can add furniture pieces using augmented reality techniques. In this thesis we perform an experimental study of the state-of-the-art registration methods to find which one fits better to our scene mapping purposes. Different methods are tested and analyzed on different scene distributions of visual and geometry appearance. In addition, this thesis proposes two methods for 3d data compression and representation of 3D maps. Our 3D representation proposal is based on the use of Growing Neural Gas (GNG) method. This Self-Organizing Maps (SOMs) has been successfully used for clustering, pattern recognition and topology representation of various kind of data. Until now, Self-Organizing Maps have been primarily computed offline and their application in 3D data has mainly focused on free noise models without considering time constraints. Self-organising neural models have the ability to provide a good representation of the input space. In particular, the Growing Neural Gas (GNG) is a suitable model because of its flexibility, rapid adaptation and excellent quality of representation. However, this type of learning is time consuming, specially for high-dimensional input data. Since real applications often work under time constraints, it is necessary to adapt the learning process in order to complete it in a predefined time. This thesis proposes a hardware implementation leveraging the computing power of modern GPUs which takes advantage of a new paradigm coined as General-Purpose Computing on Graphics Processing Units (GPGPU). Our proposed geometrical 3D compression method seeks to reduce the 3D information using plane detection as basic structure to compress the data. This is due to our target environments are man-made and therefore there are a lot of points that belong to a plane surface. Our proposed method is able to get good compression results in those man-made scenarios. The detected and compressed planes can be also used in other applications as surface reconstruction or plane-based registration algorithms. Finally, we have also demonstrated the goodness of the GPU technologies getting a high performance implementation of a CAD/CAM common technique called Virtual Digitizing.

Exposition universelle d'Anvers, Antwerp, Belgium, 1885 (Raster Image)

This layer is a georeferenced raster image of the historic paper map entitled: Plan instantané de la ville d'Anvers : indiquant l'Exposition Universelle et tous les principaux monuments en relief, le tracé des tramways, les nouveaux quais hangars, etc., etc., dressé par E.V.D.H. It was published by Richd. Huybrechts & Cie in 1885. Scale [ca. 1:11,000]. Covers Antwerp, Belgium. Map in French. The image inside the map neatline is georeferenced to the surface of the earth and fit to the Belgian Lambert 1972 coordinate system. All map collar and inset information is also available as part of the raster image, including any inset maps, profiles, statistical tables, directories, text, illustrations, index maps, legends, or other information associated with the principal map. This map shows features such as roads, railroads, drainage, built-up areas and selected buildings, ground cover, fortifications, canals, docks, and more. Relief shown pictorially. Shows buildings and points of interest related to the Exposition universelle d'Anvers.This layer is part of a selection of digitally scanned and georeferenced historic maps from the Harvard Map Collection. These maps typically portray both natural and manmade features. The selection represents a range of originators, ground condition dates, scales, and map purposes.

Washington, D.C., 1791 (Raster Image)

This layer is a georeferenced raster image of the historic paper map entitled: Plan of the city intended for the permanent seat of the government of t[he] United States : projected agreeable to the direction of the President of the United States, in pursuance of an act of Congress passed the sixteenth day of July, MDCCXC, "establishing the permanent seat on the bank of the Potowmac", by Peter Charles L'Enfant. It was published by U.S. Coast and Geodetic Survey in 1887. Scale ca. 1:15,840. Facsimile of L'Enfant's 1791 plan for Washington, D.C. The image inside the map neatline is georeferenced to the surface of the earth and fit to the Maryland State Plane Coordinate System Meters NAD83 (Fipszone 1900). All map collar and inset information is also available as part of the raster image, including any inset maps, profiles, statistical tables, directories, text, illustrations, index maps, legends, or other information associated with the principal map. This map shows features such as proposed streets and government building locations, drainage, parks, and more. Includes notes, text, statistical data, and index to points of interest. This layer is part of a selection of digitally scanned and georeferenced historic maps from The Harvard Map Collection as part of the Imaging the Urban Environment project. Maps selected for this project represent major urban areas and cities of the world, at various time periods. These maps typically portray both natural and manmade features at a large scale. The selection represents a range of regions, originators, ground condition dates, scales, and purposes.

Washington, D.C., 1792 (Raster Image)

This layer is a georeferenced raster image of the historic paper map entitled: Plan of the city of Washington in the territory of Columbia : ceded by the states of Virginia and Maryland to the United States of America, and by them established as the seat of their government, after the year MDCCC, engrav'd by Sam'l Hill, Boston ; in order to execute this plan, Mr. Ellicott drew a true meridional line. Andrew Ellicott plan of Washington, D.C. It was published in 1792. Scale [ca. 1:19,800]. The image inside the map neatline is georeferenced to the surface of the earth and fit to the Maryland State Plane Coordinate System Meters NAD83 (Fipszone 1900). All map collar and inset information is also available as part of the raster image, including any inset maps, profiles, statistical tables, directories, text, illustrations, index maps, legends, or other information associated with the principal map. This map shows features such as roads, block numbers and proposed government buildings, drainage, and more. Relief is shown by hachures. Depths shown by soundings. Includes text and notes. This layer is part of a selection of digitally scanned and georeferenced historic maps from The Harvard Map Collection as part of the Imaging the Urban Environment project. Maps selected for this project represent major urban areas and cities of the world, at various time periods. These maps typically portray both natural and manmade features at a large scale. The selection represents a range of regions, originators, ground condition dates, scales, and purposes.

Washington, D.C., 1792 (Raster Image)

This layer is a georeferenced raster image of the historic paper map entitled: Plan of the city of Washington in the territory of Columbia : ceded by the states of Virginia and Maryland to the United States of America, and by them established as the seat of their government, after the year MDCCC, engrav'd by Thackara & Vallance Philad'a 1792 ; in order to execute this plan, Mr. Ellicott drew a true meridional line ... and left nothing to the uncertainty of the compass. Andrew Ellicott plan of Washington, D.C. It was published in 1792. Scale [ca. 1:19,800]. The image inside the map neatline is georeferenced to the surface of the earth and fit to the Maryland State Plane Coordinate System Meters NAD83 (Fipszone 1900). All map collar and inset information is also available as part of the raster image, including any inset maps, profiles, statistical tables, directories, text, illustrations, index maps, legends, or other information associated with the principal map. This map shows features such as roads, block numbers and proposed government buildings, drainage, and more. Relief is shown by hachures. Depths are shown by soundings. Includes text, notes, and coat-of-arms. This layer is part of a selection of digitally scanned and georeferenced historic maps from The Harvard Map Collection as part of the Imaging the Urban Environment project. Maps selected for this project represent major urban areas and cities of the world, at various time periods. These maps typically portray both natural and manmade features at a large scale. The selection represents a range of regions, originators, ground condition dates, scales, and purposes.

Washington, D.C., 1792 (Raster Image)

This layer is a georeferenced raster image of the historic paper map entitled: Plan of the city of Washington, Thackara & Vallance, sc. Andrew Ellicott plan of Washington, D.C. It was published in the Universal asylum, and Columbian magazine, v. 4, March 1792. Scale not given. The image inside the map neatline is georeferenced to the surface of the earth and fit to the Maryland State Plane Coordinate System Meters NAD83 (Fipszone 1900). All map collar and inset information is also available as part of the raster image, including any inset maps, profiles, statistical tables, directories, text, illustrations, index maps, legends, or other information associated with the principal map. This map shows features such as proposed streets and government building locations, drainage, parks, and more. Relief is shown by hachures. This layer is part of a selection of digitally scanned and georeferenced historic maps from The Harvard Map Collection as part of the Imaging the Urban Environment project. Maps selected for this project represent major urban areas and cities of the world, at various time periods. These maps typically portray both natural and manmade features at a large scale. The selection represents a range of regions, originators, ground condition dates, scales, and purposes.

Washington, D.C., 1795 (Raster Image)

This layer is a georeferenced raster image of the historic paper map entitled: Plan of the city of Washington, in the territory of Columbia, ceded by the States of Virginia and Maryland to the United States of America, and by them established as the seat of their government after the year 1800, J. Russell, sculpt., Constitu'n Row, Grays Inn Road. It was published in 1795. Scale [ca 1:19,800]. The image inside the map neatline is georeferenced to the surface of the earth and fit to the Maryland State Plane Coordinate System Meters NAD83 (Fipszone 1900). All map collar and inset information is also available as part of the raster image, including any inset maps, profiles, statistical tables, directories, text, illustrations, index maps, legends, or other information associated with the principal map. This map shows features such as roads, block numbers and proposed government buildings, drainage, and more. Relief shown by hachures. Depths shown by soundings. Includes "Remarks" and coat-of-arms. This layer is part of a selection of digitally scanned and georeferenced historic maps from The Harvard Map Collection as part of the Imaging the Urban Environment project. Maps selected for this project represent major urban areas and cities of the world, at various time periods. These maps typically portray both natural and manmade features at a large scale. The selection represents a range of regions, originators, ground condition dates, scales, and purposes.

Washington, D.C., 1815 (Raster Image)

This layer is a georeferenced raster image of the historic paper map entitled: Territory of Columbia, drawn by And'w Ellicott ; engraved by P.A.F. Tardieu, Paris, 1815. It was published in 1815. Scale [ca. 1:63,360]. The image inside the map neatline is georeferenced to the surface of the earth and fit to the Maryland State Plane Coordinate System Meters NAD83 (Fipszone 1900). All map collar and inset information is also available as part of the raster image, including any inset maps, profiles, statistical tables, directories, text, illustrations, index maps, legends, or other information associated with the principal map. This map shows features such as roads, drainage, district boundaries, government buildings, parks and more. Relief is shown by hachures. This layer is part of a selection of digitally scanned and georeferenced historic maps from The Harvard Map Collection as part of the Imaging the Urban Environment project. Maps selected for this project represent major urban areas and cities of the world, at various time periods. These maps typically portray both natural and manmade features at a large scale. The selection represents a range of regions, originators, ground condition dates, scales, and purposes.

Washington, D.C., 1820 (Raster Image)

This layer is a georeferenced raster image of the historic paper map entitled: A correct map of the city of Washington : capital of the United States of America : lat. 38.53 n., long. 0.0, eng'd by W.I. Stone, Wash'n. It was published by Davis & Force in 1820. Scale [ca. 1:19,800]. The image inside the map neatline is georeferenced to the surface of the earth and fit to the Maryland State Plane Coordinate System Meters NAD83 (Fipszone 1900). All map collar and inset information is also available as part of the raster image, including any inset maps, profiles, statistical tables, directories, text, illustrations, index maps, legends, or other information associated with the principal map. This map shows features such as roads, drainage, selected government and public buildings, block numbers. churches, and more. Includes ill. and index to points of interest. This layer is part of a selection of digitally scanned and georeferenced historic maps from The Harvard Map Collection as part of the Imaging the Urban Environment project. Maps selected for this project represent major urban areas and cities of the world, at various time periods. These maps typically portray both natural and manmade features at a large scale. The selection represents a range of regions, originators, ground condition dates, scales, and purposes.

Washington, D.C., 1839 (Raster Image)

This layer is a georeferenced raster image of the historic paper map entitled: Map of the city of Washington in the District of Columbia : established as the permanent seat of the government of the United States of America, W.J. Stone sc. Wash'n. It was published ca. 1839. Scale [ca. 1:12,250]. The image inside the map neatline is georeferenced to the surface of the earth and fit to the Maryland State Plane Coordinate System Meters NAD83 (Fipszone 1900). All map collar and inset information is also available as part of the raster image, including any inset maps, profiles, statistical tables, directories, text, illustrations, index maps, legends, or other information associated with the principal map. This map shows features such as roads, drainage, block numbers, wards, selected government buildings, and more. Relief is shown by hachures. Includes indexed table of "References." This layer is part of a selection of digitally scanned and georeferenced historic maps from The Harvard Map Collection as part of the Imaging the Urban Environment project. Maps selected for this project represent major urban areas and cities of the world, at various time periods. These maps typically portray both natural and manmade features at a large scale. The selection represents a range of regions, originators, ground condition dates, scales, and purposes.

Washington, D.C., 1840 (Raster Image)

This layer is a georeferenced raster image of the historic paper map entitled: Map of the city of Washington, drawn by F.C. De Krafft, city surveyor ; eng'd by Mrs. W.I. Stone. It was published by Wm. M. Morrison in 1840. Scale [ca. 1:18,400]. The image inside the map neatline is georeferenced to the surface of the earth and fit to the Maryland State Plane Coordinate System Meters NAD83 (Fipszone 1900). All map collar and inset information is also available as part of the raster image, including any inset maps, profiles, statistical tables, directories, text, illustrations, index maps, legends, or other information associated with the principal map. This map shows features such as roads, drainage, block numbers, city wards, selected government buildings, and more. This layer is part of a selection of digitally scanned and georeferenced historic maps from The Harvard Map Collection as part of the Imaging the Urban Environment project. Maps selected for this project represent major urban areas and cities of the world, at various time periods. These maps typically portray both natural and manmade features at a large scale. The selection represents a range of regions, originators, ground condition dates, scales, and purposes.

Washington, D.C., 1855 (Raster Image)

This layer is a georeferenced raster image of the historic paper map entitled: Colton's Georgetown and the city of Washington : the capital of the United States of America. It was published by J.H. Colton in 1855. Scale [ca. 1:25,750]. The image inside the map neatline is georeferenced to the surface of the earth and fit to the Maryland State Plane Coordinate System Meters NAD83 (Fipszone 1900). All map collar and inset information is also available as part of the raster image, including any inset maps, profiles, statistical tables, directories, text, illustrations, index maps, legends, or other information associated with the principal map. This map shows features such as roads, drainage, block numbers, city wards, built-up areas, selected government buildings, parks, and more. Includes views: Smithsonian Institution -- The Capitol -- Washington Monument. This layer is part of a selection of digitally scanned and georeferenced historic maps from The Harvard Map Collection as part of the Imaging the Urban Environment project. Maps selected for this project represent major urban areas and cities of the world, at various time periods. These maps typically portray both natural and manmade features at a large scale. The selection represents a range of regions, originators, ground condition dates, scales, and purposes.

Washington, D.C., 1884 (Raster Image)

This layer is a georeferenced raster image of the historic paper map entitled: Topographical map of the District of Columbia and a portion of Virginia, compiled under the direction of Major G.J. Lydecker, Corps of Engineers, Engineer Commissioner D.C., by Captain F.V. Greene, Corps of Engineers ; drawn by W.T.O. Bruff. It was published in 1884. Scale [1:15,840]. The image inside the map neatline is georeferenced to the surface of the earth and fit to the Maryland State Plane Coordinate System Meters NAD83 (Fipszone 1900). All map collar and inset information is also available as part of the raster image, including any inset maps, profiles, statistical tables, directories, text, illustrations, index maps, legends, or other information associated with the principal map. This map shows features such as roads, railroads, drainage, selected public buildings, selected private residences and names of landowners, built-up areas, parks, and more. Relief shown by contours. Depths shown by soundings. Includes source materials note. This layer is part of a selection of digitally scanned and georeferenced historic maps from The Harvard Map Collection as part of the Imaging the Urban Environment project. Maps selected for this project represent major urban areas and cities of the world, at various time periods. These maps typically portray both natural and manmade features at a large scale. The selection represents a range of regions, originators, ground condition dates, scales, and purposes.

Washington, D.C., 1895 (Raster Image)

This layer is a georeferenced raster image of the historic paper map entitled: Rand, McNally & Co.'s indexed atlas of the world : map of Washington, D.C. It was published by Rand, McNally & Co. in 1895. Scale [ca. 1:15,000]. The image inside the map neatline is georeferenced to the surface of the earth and fit to the Maryland State Plane Coordinate System Meters NAD83 (Fipszone 1900). All map collar and inset information is also available as part of the raster image, including any inset maps, profiles, statistical tables, directories, text, illustrations, index maps, legends, or other information associated with the principal map. This map shows features such as roads, railroads, drainage, selected public buildings and points of interest, parks, and more. This layer is part of a selection of digitally scanned and georeferenced historic maps from The Harvard Map Collection as part of the Imaging the Urban Environment project. Maps selected for this project represent major urban areas and cities of the world, at various time periods. These maps typically portray both natural and manmade features at a large scale. The selection represents a range of regions, originators, ground condition dates, scales, and purposes.

Washington, D.C., 1933 (Raster Image)

This layer is a georeferenced raster image of the historic paper map entitled: Base map of the District of Columbia showing railroads, main thoroughfares, alley blocks, and public (government) buildings, base prepared in the Office of the Surveyor, D.C., by direction of the Engineer Commissioner, D.C. It was published in 1936. Scale [ca. 1:19,000]. Base map "Complete to June 13, 1933." The image inside the map neatline is georeferenced to the surface of the earth and fit to the Maryland State Plane Coordinate System Meters NAD83 (Fipszone 1900). All map collar and inset information is also available as part of the raster image, including any inset maps, profiles, statistical tables, directories, text, illustrations, index maps, legends, or other information associated with the principal map. This map shows features such as roads, railroads and stations, drainage, selected public buildings and points of interest, parks, cemeteries, and more. Also shows block numbers and proposed streets. This layer is part of a selection of digitally scanned and georeferenced historic maps from The Harvard Map Collection as part of the Imaging the Urban Environment project. Maps selected for this project represent major urban areas and cities of the world, at various time periods. These maps typically portray both natural and manmade features at a large scale. The selection represents a range of regions, originators, ground condition dates, scales, and purposes.