Aniello Panariello

PhD Student in Artificial Intelligence

AImageLab @ University of Modena and Reggio Emilia

About Me

I am Aniello Panariello, a Ph.D. student affiliated with the AImageLab research group at the University of Modena and Reggio Emilia, under the guidance of Prof. Simone Calderara, I am immersed in exploring Computer Vision and Deep Learning techniques. My prior experience includes leading the Computer Division for the Formula Student Team - MMR Driverless, where I focused on enhancing the perception capabilities of an autonomous racing car.

Two primary focal points characterize my research journey. The first centers around Monocular Distance Estimation, a challenging area where I explore methods to accurately gauge distances using single-camera inputs. The second area of my research revolves around Video Anomaly Detection, a field focused on identifying unusual human activities within video clips. In both domains, I employ innovative unsupervised and weakly supervised methods to maximize the utility of limited annotations while contributing to advancing these cutting-edge technologies.

Feel free to reach me for any question or curiosity.

Interests

Computer Vision
Machine Learning
Deep Learning
Video Anomaly Detection
Self-supervised Learning
Continual Learning

Education

PhD in Artificial Intelligence, 2025

University of Modena and Reggio Emilia
MS in Artificial Intelligence, 2021

University of Modena and Reggio Emilia
BSc in Computer Engineering, 2019

University of Salerno

Experience

Research Fellow

AImageLab @ University of Modena and Reggio Emilia

Jan 2022 – Oct 2022 Modena, Italy

Developed a deep learning based software for the detection of video anomalous activities on public transport for the pan-European project InSecTT.
Studied and investigated techniques for data imbalance mitigation in anomaly detection problems.
Introduced a novel consistency-based regulation loss for temporal anomaly localization in self-supervised learning settings.
Developing of new techniques for video anomaly detection and video explaination using VAEs.

Computer Division Leader

MMR Driverless - Formula Student Team

Nov 2019 – Jul 2022 Modena, Italy

Construction of a dataset of traffic cones consisting of four classes and 9000 images, with which a neural network has been fine-tuned to perform cones detection in images.
Developing of a deep learning based software for autonomous car perception based on cameras and Lidar sensors. The car is able to accurately detect the color and precise distance of traffic cones that define the track, in a 180° FOV ahead.
Project manager for the computer division.

Intern

Mivia Lab @ University of Salerno

Dec 2018 – Mar 2019 Salerno, Italy

Developed “ARPosture” (Swift) for the thesis work. The app was presented to the head of the Apple Developer Academy, being among the top 5 best apps developed at the University of Salerno in 2017-2019.

iOS App Developer

Apple Foundation Program @ University of Salerno

Jun 2018 – Jul 2018 Salerno, Italy

Participated in the iOS Foundation Program, which improved my team-working, team-leading and app developing skills.
Was the lead developer and project manager of the “Kelnero” app (Swift).

News

[Oct. 2023] I participated in the ICCV2023 conference in Paris, France.
[Sep. 2023] I participated in the ELLIS Summer School on Large-Scale AI for Research and Industry in Modena, Italy.
[Sep. 2023] I participated in the VISMAC 23 Summer School in Padova, Italy.
[Jul. 2023] Our paper TrackFlow: Multi-Object tracking with Normalizing Flows has been accepted at ICCV 2023.
[Nov. 2022] I started my Ph.D. in Information and Communication Technologies (ICT) at AImageLab, with topic “AI techniques for time series analysis and prediction exploiting structured information”. (Granted by Ammagamma)
[Aug. 2022] Our paper Consistency-based Self-supervised Learning for Temporal Anomaly Localization has been accepted at ECCV2022 (Workshops).
[Jan. 2022] I joined the AImageLab group at UNIMORE in Italy as a Research Fellow.

Featured Publications

TrackFlow: Multi-Object tracking with Normalizing Flows

We propose a novel approach to multi-object tracking that leverages Normalizing Flows to learn a joint probability distribution over the costs of candidate associations. Our experiments show that our approach consistently enhances the performance of several tracking-by-detection algorithms.

Gianluca Mancusi, Aniello Panariello, Angelo Porrello, Matteo Fabbri, Simone Calderara, Rita Cucchiara

Consistency-based Self-supervised Learning for Temporal Anomaly Localization

This work tackles Weakly Supervised Anomaly detection within the field of self-supervised learning, asking the model to yield the same anomaly scores for different augmentations of the same video sequence.

Aniello Panariello, Angelo Porrello, Simone Calderara, Rita Cucchiara

Recent Publications

Check my Google Scholar profile for the complete publication list.

Quickly discover relevant content by filtering publications.

Gianluca Mancusi, Aniello Panariello, Angelo Porrello, Matteo Fabbri, Simone Calderara, Rita Cucchiara (2023). TrackFlow: Multi-Object tracking with Normalizing Flows. In ICCV'23.

PDF Cite Poster

Aniello Panariello, Angelo Porrello, Simone Calderara, Rita Cucchiara (2022). Consistency-based Self-supervised Learning for Temporal Anomaly Localization. In ECCVW'22.

PDF Cite Code Slides DOI