GSoC2011SfM
0.1
Google Summer of Code 2011: Structure from motion
|
OpencvSfM::bundle_datas | |
OpencvSfM::Camera | This class represent the physical device which take the pictures. It is not related to a 3D position which is the role of the PointOfView class. The role of the class is to store only device related informations like intra parameters, radial and tangential distotion. This abstract class is not related to a type of camera ( fish eyes... ) |
OpencvSfM::CameraPinhole | This class represent the physical device which take the pictures. It is not related to a 3D position which is the role of the PointOfView class. The role of the class is to store only intra parameters ( without radial distortion ) |
OpencvSfM::CameraPinholeDistor | This class represent the physical device which take the pictures. It is not related to a 3D position which is the role of the PointOfView class. The role of the class is to store intra parameters and radial distortion |
OpencvSfM::EuclideanEstimator | This class perform a projective estimation of the motion. Given points matches and cameras with intra parameters, it tries to find the best cameras positions and 3D points. Does not perform a bundle ajustement! |
OpencvSfM::ImageLink | This structure store an image link ( two image ids ).. |
OpencvSfM::ImagesGraphConnection | This class modelizes the images graph connections |
OpencvSfM::MatcherSparseFlow | A class used for matching points between two images |
OpencvSfM::MatchingThread | This struct is used by boost::thread object to compute match. I used some semaphore to ensure the matching process work well |
OpencvSfM::Mesh | This class is used to handle a entire mesh |
OpencvSfM::MotionProcessor | This class try to create a commun interface for files loading. Indeed, if you want to use webcam, avi file of list of files, you will have to do some annoying processing, like iterate the different files of the directory. With MotionProcessor, you can now use a folder of image the same way you use a webcam or a video file |
OpencvSfM::mapping::Point | This structure will handle conversions between OpenCV and PCL data |
OpencvSfM::PointOfView | This class represent the 3D position of the device which take the pictures. The role of the class is to store everything related to the filed of view: picture, 3D position, points, matches and 3D points |
OpencvSfM::PointsMatcher | A class used for matching descriptors that can be described as vectors in a finite-dimensional space |
OpencvSfM::PointsMatcherOpticalFlow | A class used for matching points between two images |
OpencvSfM::PointsToTrack | This class can be used to store informations about point and features. This is an abstract class: you can't use it directly. Use for instance PointsToTrackWithImage |
OpencvSfM::PointsToTrackWithImage | This class can be used to find points and features in pictures using SIFT detector |
OpencvSfM::Polygon | This class is used to handle a facet of a mesh |
OpencvSfM::SequenceAnalyzer | This class tries to match points in the entire sequence. It follow ideas proposed by Noah Snavely: Modeling the World from Internet Photo Collections |
OpencvSfM::StructureEstimator | This class tries to find the 3D structure using a sequence and cameras fully parameterized |
OpencvSfM::TrackOfPoints | This class store the track of keypoints. A track is a connected set of matching keypoints across multiple images |
OpencvSfM::Visualizer | This class can be used to view the differents object involved in current structure from motion process |