Cosimo Rubino received the PhD in Computer Vision
from the University of Genova, while he worked as a fellow researcher in Visual Geometry and Modelling (VGM) Lab at
the PAVIS department of the Istituto Italiano di Tecnologia (IIT) in Genova. He received the BSc degree in
Industrial Engineering and the MSc degree in Mechatronic Engineering in 2012 from University of Trento.
His research focuses in the areas of 3D scene understanding and motion segmentation.
PhD in Computer Vision
MSc in Mechatronic Engineering
BSc in Industrial Engineering
Thesis Title: Implementation and comparison among vehicles driving models
This paper presents a method to deal with the multi-body segmentation problem using a set of 2D points matches between two views. The key feature of our approach is the explicit inclusion of a higher semantic information as given by general purpose object detectors that boost the segmentation of the moving objects.Multi-view geometry, Motion Segmentation
This work shows how to reconstruct the position of rigid objects and to jointly recover affine camera calibration solely from a set of object detections in a video sequence. In practice, this work can be considered as the extension of Tomasi and Kanade factorization method using objects. Instead of using points to form a rank constrained measurement matrix, we can form a matrix with similar rank properties using 2D object detection proposals.3D Object Localisation, 3D Reconstruction
In this work is presented a novel approach to recover objects 3D position and occupancy in a generic scene using only 2D object detections from multiple view images. The method reformulates the problem as the estimation of a quadric (ellipsoid) in 3D given a set of 2D bounding box detections in multiple views. We show that a closed-form solution exists using a minimum of three views, while a solution with two views is possible through the use of non-linear optimisation and object constraints on the size of the object shape.3D Object Localisation, 3D Reconstruction
In this work are presented two new methods based on Interval Analysis and on Computational Geometry for estimating the 3D occupancy and position of objects from image sequences. Given a calibrated set of images, the proposed framework frst detect objects using off-the-shelf object detectors and then match bounding boxes in multiple views. Then, the 2D image constraints given by the bounding boxes are used to effciently recover 3D object position and occupancy using solely geometrical constraints in multiple views.3D Object Localisation, 3D Reconstruction
This thesis proposes novel approaches for scene understanding using RGB images, in particular for the motion segmentation and the object localisation problems. For segmenting motions two novel frameworks are described: A pair-wise consensus and a n-view optimisation based approaches. The object localisation task is performed by a multi-view technique, which handles the information provided by the object detector through the bounding boxes in order to estimate the volume occupied by the objects. The method is geometric and has been formulated in closed form for both the perspective and orthographic camera models.Motion Segmentation, 3D Object Localisation, 3D Reconstruction
This paper presents a new approach for image-based motion segmentation in the case of vehicles navigating inside an urban environment. By constraining the geometry and exploiting known semantic classes in the scene, we achieve much higher accuracy than previous approaches.Motion Segmentation, Semantics