Pietro Cerri

Publications : 40
Aldex : 30
H-index : 19
Citations : 1050

CoMoGAN: continuous model-guided image-to-image translation

Fabio Pizzati et 3 al.

Mar 11, 2021 in Arxiv
CoMoGAN is a continuous GAN relying on the unsupervised reorganization of the target data on a functional manifold. To that matter, we introduce a new Functional Instance Normalization layer and residual mechanism, which together disentangle image content from position on target manifold. We rely on naive physics-inspired models to guide the training while allowing private model/translations featu...

Physics-informed Guided Disentanglement in Generative Networks

Fabio Pizzati et 3 al.

Jul 29, 2021 in Arxiv
Image-to-image translation (i2i) networks suffer from entanglement effects in presence of physics-related phenomena in target domain (such as occlusions, fog, etc), lowering altogether the translation quality, controllability and variability. In this paper, we build upon collection of simple physics models and present a comprehensive method for disentangling visual traits in target images, guiding...

Physics-informed Guided Disentanglement in Generative Networks

Fabio Pizzati et 3 al.

Jul 29, 2021 in Arxiv
Image-to-image translation (i2i) networks suffer from entanglement effects in presence of physics-related phenomena in target domain (such as occlusions, fog, etc), lowering altogether the translation quality, controllability and variability. In this paper, we build upon collection of simple physics models and present a comprehensive method for disentangling visual traits in target images, guiding...

engineering

Sensing requirements for a 13,000 km intercontinental autonomous drive

Alberto Broggi et 5 al.

Jan 1, 2010 in 2010 IEEE Intelligent Vehicles Symposium
This paper presents the design issues that were considered for the equipment of 4 identical autonomous vehicles that will drive themselves without human intervention on an intercontinental route for more than 13,000 km.

Model-Based Occlusion Disentanglement for Image-to-Image Translation

Fabio Pizzati et 3 al.

Jan 1, 2020
Image-to-image translation is affected by entanglement phenomena, which may occur in case of target data encompassing occlusions such as raindrops, dirt, etc. Our unsupervised model-based learning disentangles scene and occlusions, while benefiting from an adversarial pipeline to regress physical parameters of the occlusion model. The experiments demonstrate our method is able to handle varying ty...

engineering

Computer vision at the hyundai autonomous challenge

Pietro Cerri et 8 al.

Jan 1, 2011 in 2011 14th International IEEE Conference on Intelligent Transportation Systems
As interest on autonomous vehicles is growing worldwide, different approaches, based on different perception technologies and concepts, are being followed. This paper exposes the importance of the use of vision technology in most of these approaches, and presents the experience of the SNUCLE autonomous vehicle which successfully completed the Hyundai Autonomous Challenge in November 2010.

Vision Technologies for Intelligent Vehicles

Massimo Bertozzi, Paolo Zani et 15 al.

Jan 1, 2007
This paper surveys the different technologies involved in the development of computer vision applications for Intelligent Vehicles derived from the 15 year experience of VisLab. Some illustrative examples are also discussed.

engineering

An embedded system for counting passengers in public transportation vehicles

Nicola Bernini et 5 al.

Jan 1, 2014 in 2014 IEEE/ASME 10th International Conference on Mechatronic and Embedded Systems and Applications
This article describes a system for people counting conceived for public transportation vehicles. The underlying idea is to monitor the number of passengers getting in or out public transportation means like buses and metros over time hence computing reliable estimations in order to improve vehicle`s door control. A stereo vision system is presented, it has been developed considering its future in...

Suspended traffic lights detection and distance estimation using color features

Moisés Díaz Cabrera et 3 al.

Jan 1, 2012 in 2012 15th International IEEE Conference on Intelligent Transportation Systems
Traffic Light Detection is a problem differently approached by many research groups around the world. Here we present a novel technique to detect suspended traffic lights, based on colors and features such as black area of traffic lights or area of lighting lamps. Additionally, the traffic light distance is estimated aiming at slowing down and stopping in the correct position, in case of red light...

VIAC: An out of ordinary experiment

Massimo Bertozzi et 22 al.

Jan 1, 2011 in 2011 IEEE Intelligent Vehicles Symposium
This paper presents the preliminary results of VIAC, the VisLab Intercontinental Autonomous Challenge, a test of autonomous driving along an unknown route from Italy to China. It took 3 months to run the entire test; all data have been logged, including all data generated by the sensors, vehicle data, and GPS info. This huge amount of information has been packed during the trip, compressed, and tr...

Similar author:

Danyang Tian