Simone Milani | Associate Professor

Phone (office): +39 049 827 7641

Biography

Simone Milani was born in 1978. He received his Laurea Degree (5 years course) with honors from the University of Padova, Italy, on December 2002 and the Ph.D. degree in Electronics and Telecommunication Engineering on March 2007. From August 2005 to June 2006 he was visiting Ph.D. student at the University of California – Berkeley, Berkeley, CA, USA. From January 2007 he started working as a post-doc researcher for the University of Padova, University of Udine and Politecnico di Milano, Italy. On May 2014 he started working as assistant professor at the University of Padova until March 2020 since he got an associate professor at the same university. He has also been a visiting researcher at the Universidade Estadual de Campinas, Brasil in 2019. He works in the multimedia research group and his research activity is focused on image/video/audio/3D data compression and multimedia forensics.

His research interest includes 3D reconstruction and understanding, light field data processing, learned compression scheme, point cloud compression/segmentation/classification, semantic segmentation of 3D data, audio compression, Time-of-Flight sensor data processing, depth compression, multimedia forensics, biometrics, anomaly detection and multimedia security issues. He is an associate editor of IEEE Transactions on Multimedia and IEEE Transaction on Image Processing. He has been involved in several European (FP7 REWIND, POR FESR Rebuilding the Past), National and university funded projects. He has co-authored many journal papers and conference proceedings.

Research areas

Point cloud compression

Multi-Modal learning

Hand Gesture Recognition

Depth Estimation

Continual learning

Publications

136 entries « ‹ 1 of 3 › »

2024

Camuffo, Elena; Michieli, Umberto; Milani, Simone; Moon, Ji Joong; Ozay, Mete

Enhanced Model Robustness to Input Corruptions by Per-corruption Adaptation of Normalization Statistics Proceedings Article

In: IEEE, (Ed.): IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2024.

Abstract | Links | BibTeX

@inproceedings{nokey,

title = {Enhanced Model Robustness to Input Corruptions by Per-corruption Adaptation of Normalization Statistics},

author = {Elena Camuffo and Umberto Michieli and Simone Milani and Ji Joong Moon and Mete Ozay},

editor = {IEEE},

url = {https://arxiv.org/abs/2407.06450},

year  = {2024},

date = {2024-06-30},

urldate = {2024-06-30},

booktitle = { IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},

journal = {IEEE},

abstract = {Developing a reliable vision system is a fundamental challenge for robotic technologies (e.g., indoor service robots and outdoor autonomous robots) which can ensure reliable navigation even in challenging environments such as adverse weather conditions (e.g., fog, rain), poor lighting conditions (e.g., over/under exposure), or sensor degradation (e.g., blurring, noise), and can guarantee high performance in safety-critical functions. Current solutions proposed to improve model robustness usually rely on generic data augmentation techniques or employ costly test-time adaptation methods. In addition, most approaches focus on addressing a single vision task (typically, image recognition) utilising synthetic data. In this paper, we introduce Per-corruption Adaptation of Normalization statistics (PAN) to enhance the model robustness of vision systems. Our approach entails three key components: (i) a corruption type identification module, (ii) dynamic adjustment of normalization layer statistics based on identified corruption type, and (iii) real-time update of these statistics according to input data. PAN can integrate seamlessly with any convolutional model for enhanced accuracy in several robot vision tasks. In our experiments, PAN obtains robust performance improvement on challenging real-world corrupted image datasets (e.g., OpenLoris, ExDark, ACDC), where most of the current solutions tend to fail. Moreover, PAN outperforms the baseline models by 20-30% on synthetic benchmarks in object recognition tasks.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}