Andrea Romanoni

PhD

About me

I am a Post-Doc at Politecnico di Milano. My research is focused on Computer Vision, especially 3D Reconstruction. During my Bachelor and my Master theses, I developed a visual vehicle tracking system, that reconstructs the trajectory of the vehicles both on the image and in the world space. This project involved different topics: objects detection with the Background Subtraction, data association, Kalman and Particle Filtering and Smoothing.

During my PhD, I moved to visual 3D reconstruction, and I proposed a completely automatic pipeline that models, incrementally, the scene acquired from the images. The pipeline adopts a manifold reconstruction from Structure from Motion point cloud method, a novel mesh merging algorithm and photometric surface evolution.

Keeping my focous on 3D reconstruction and Multi-View Stereo, I am now investigating how to improve the entire MVS pipeline, from depth map estimation to mesh refinement by combining classical approaches, such as PatchMatch, with machine learning oriented method, such as geometric deep learning.

Publications

Conferences

2019

  • A. Romanoni, M. Matteucci. TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. International Conference on Computer Vision (ICCV) 2019. PDF BibTex
  • A. Romanoni, M. Matteucci. Dense 3D Visual Mapping via Semantic Simplification. International Conference on Robotics and Automation (ICRA) 2019. PDF BibTex
  • A. Romanoni, M. Matteucci. Asynchronous Convolutional Networks for Object Detection in Neuromorphic Cameras. Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) 2019 (Best Paper Award). PDF BibTex
  • A. Romanoni, M. Matteucci. Attention Mechanisms for Object Recognition with Event-Based Cameras. Winter Conference on Applications of Computer Vision (WACV) 2019. PDF BibTex

2018

  • A. Romanoni, M. Matteucci. A Data-driven Prior on Facet Orientation for Semantic Mesh Labeling. International Conference on 3D Vision (3DV) 2018. PDF BibTex
  • A. Bignoli, A. Romanoni, M. Matteucci. Real-Time CPU-Based Large-Scale Three-Dimensional Mesh Reconstruction. IEEE Robotics and Automation Letters 2018. PDF BibTex
  • L. Morreale, A. Romanoni, M. Matteucci. Predicting the Next Best View for 3D Mesh Refinement. Intelligent Autonomous Systems 15 (IAS-15) 2018. PDF BibTex
  • A. Bignoli, A. Romanoni, M. Matteucci. Multi-View Stereo 3D Edge Reconstruction. IEEE Winter Conference on Applications of Computer Vision (WACV) 2018. PDF BibTex

2017

  • A. Romanoni, D. Fiorenti, M. Matteucci. Mesh-based 3D textured urban mapping. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2017. PDF BibTex
  • A. Romanoni, M. Ciccone, F. Visin M. Matteucci. Multi-view stereo with single-view semantic mesh refinement. IEEE International Conference on Computer Vision Workshops (ICCVW) 2017 PDF BibTex

2016

  • G. Postica, A. Romanoni, M. Matteucci. Robust Moving Objects Detection in Lidar Data Exploiting Visual Cues. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2016. PDF BibTex
  • A. Romanoni, A. Delaunoy, M. Pollefeys, M. Matteucci. Automatic 3D Reconstruction of Manifold Meshes via Delaunay Triangulation and Mesh Sweeping. Winter Applications of Computer Vision (WACV) 2016. (Best Paper Award) PDF Video BibTex

2015

  • A. Romanoni, M. Matteucci. Incremental Urban Manifold Reconstruction from a Sparse Edge-Points Cloud. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2015. Hamburg September, 28 - October, 2, 2015 PDF Video BibTex
  • A. Romanoni, M. Matteucci. Efficient moving point handling for incremental 3D manifold reconstruction. International Conference on Image Analysis and Processing (ICIAP) 2015. Genova September, 7-11, 2015. 489-499. PDF Video BibTex

2013

  • M. Matteucci, D.Rizzi, A. Romanoni, L. Mussone. 3D Image Processing of vehicular trajectories in roundabouts. 13th World Conference of Transportation Research (WCTR), Rio de Janeiro, July 15-18, 2013. PDF BibTex

Journals

2019

  • A. Romanoni, M. Matteucci. Mesh-based camera pairs selection and occlusion-aware masking for mesh refinement. Pattern Recognition Letters (PRL). PDF BibTex

2018

  • A. Romanoni, M. Matteucci. Real-Time CPU-Based Large-Scale Three-Dimensional Mesh Reconstruction. IEEE International Conference on Robotics and Automation (ICRA). PDF BibTex

2015

  • A. Romanoni, D. G. Sorrenti, M. Matteucci. Backward-Simulation Particle Smoother with a hybrid state for 3D vehicle trajectory, class and dimension simultaneous estimation. Machine Vision and Applications, April 2015, Volume 26, Issue 2-3, pp 369-385. PDF Video Video (Experiments) BibTex
  • A. Romanoni, L. Mussone, D. Rizzi, M. Matteucci. A comparison of two Monte Carlo algorithms for 3D vehicle trajectory reconstruction in roundabouts. Pattern Recognition Letters, January 2015, Volume 51, pp 79-85. PDF Video BibTex

2014

  • A. Romanoni, M. Matteucci, D. G. Sorrenti. Background subtraction by combining Temporal and Spatio-Temporal histograms in the presence of camera movement. Machine Vision and Applications, August 2014, Volume 25, Issue 6, pp 1573-1584. PDF BibTex

Others

  • M. Matteucci, D. Rizzi, A. Romanoni. 3D reconstruction of vehicular trajectories from single and multiple camera streams. Internal Report n. 10, 2012.
  • A. Romanoni. Ricostruzione 3D delle traiettorie veicolari da immagini di una o più telecamere.. Master Thesis (in italian), Politecnico di Milano 2012. BibTex

Contacts

Building 21, Floor 1, Office 21
Via Golgi 39 - 20133 - Milano, Italy
+39.02.2399.9687
andrea.romanoni@polimi.it