Ricardo Sanchez-Matilla

Ricardo Sanchez-Matilla

Posts by Collection

publications

Online multi-target tracking with strong and weak detections

Published in Proc. of 2nd Workshop on Benchmarking Multi-target Tracking: MOTChallenge 2016, Amsterdam, October 9, 2016

This paper proposes a novel use of both high- and low-confidence target detections in a Probability Hypothesis Density Particle Filter framework for online multi-target tracker.

Recommended citation: R. Sanchez-Matilla, F. Poiesi and A. Cavallaro. "Online multi-target tracking with strong and weak detection." Proc. of 2nd Workshop on Benchmarking Multi-target Tracking: MOTChallenge.

Hierarchical detection of persons in groups

Published in Signal, Image and Video Processing (SIVP), 2017

This paper proposes a method a person detector that exploits leverages the fact that people commonly appear in groups and they occlude to each other.

Recommended citation: A. García-Martín, R. Sanchez-Matilla and José M. Martínez."Hierarchical detection of persons in groups." Signal, Image and Video Processing (SIVP).

Multi-modal localization and enhancement of multiple sound sources from a Micro Aerial Vehicle

Published in Proc. of ACM Multimedia, Mountain View, USA, October 23-27, 2017

This paper introduces for first time a multi-modal algorithim to localize and enhance multiple sound sources simultaneously from a Micro Aerial Vehicle.

Recommended citation: R. Sanchez-Matilla, L. Wang and A. Cavallaro. "Multi-modal localization and enhancement of multiple sound sources from a Micro Aerial Vehicle." Proc. of ACM Multimedia.

Tracking a moving sound source from a multi-rotor drone

Published in Proc. of IEEE/RSJ Int. Conference on Intelligent Robots and Systems (IROS), Madrid, Spain, October, 1- 5, 2018

This paper proposes to track a moving sound soure from a multi-rotor drone only using sound sensing.

Recommended citation: L. Wang, R. Sanchez-Matilla and A. Cavallaro. "Tracking a moving sound source from a multi-rotor drone." Proc. of IEEE/RSJ Int. Conference on Intelligent Robots and Systems (IROS).

Confidence Intervals for Tracking Perforamance Scores

Published in Proc. of IEEE Int. Conf. on Image Processing (ICIP), Athens, Greece, October 7-10, 2018

This paper discusses the limitations of existing ground-truth annotations that are commonly annotated via semi-automatic methods, and proposes the use of confidence intervals that can be estimated directly from already-annotated datasets.

Recommended citation: R. Sanchez-Matilla and A. Cavallaro. "Confidence Intervals for Tracking Perforamance Scores." Proc. of IEEE Int. Conf. on Image Processing (ICIP).

Scene Privacy Protection

Published in Proc. of IEEE Int. Conf. on Acoustics, Speech and Signal Processing (ICASSP), Brighton, UK, May 12-17, 2019

This paper proposes a method to protect users sensitive information from undesired automatic inferences by service providers withouth compromising the utility of the information.

Recommended citation: C.Y. Li, A.S. Shamsabadi, R. Sanchez-Matilla, R. Mazzon and A. Cavallaro. "Scene Privacy Protection." Proc. of IEEE Int. Conf. on Acoustics, Speech and Signal Processing (ICASSP).

A predictor of moving objects for first-person vision

Published in Proc. of IEEE Int. Conf. on Image Processing (ICIP), Taipei, Taiwan, September 22-25, 2019

This paper proposes a method to predict accurately the location of objects of interest from a moving camera. The method allows to forecast 60% more accurately than previously existing predictors. The method is robust to frame rate deductions of up to 66% while maintaining similar accuracy than existing methods.

Recommended citation: R. Sanchez-Matilla and A. Cavallaro. "A predictor of moving objects for first-person vision." Proc. of IEEE Int. Conf. on Image Processing (ICIP).

Audio-visual sensing from a quadcopter: dataset and baselines for source localization and sound enhancement

Published in Proc. of IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), Macau, China, Nov 4-8, 2019

This paper presents an audio-visual dataset from a quadcopter for enabling evaluation of sound source localizaiton and sound enhancement.

Recommended citation: L. Wang, R. Sanchez-Matilla and A. Cavallaro. "Audio-visual sensing from a quadcopter: dataset and baselines for source localization and sound enhancement." Proc. of IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS).

Benchmark for Human-to-Robot Handovers of Unseen Containers with Unknown Filling

Published in IEEE Transactions on Robotics and Automation Letters (RA-L): Special Issue on Benchmarking Protocols for Robotic Manipulation - To Appear, 2019, 2019

This paper proposes a benchmark for dynamic human-to-robot handovers that do not rely on a motion capture system, markers, or prior knowledge of specific objects.

Recommended citation: Ricardo Sanchez-Matilla, Konstantinos Chatzilygeroudis, Apostolos Modas, Nuno Ferreira Duarte, Alessio Xompero, Pascal Frossard, Aude Billard, and Andrea Cavallaro. "Benchmark for Human-to-Robot Handovers of Unseen Containers with Unknown Filling." IEEE Transactions on Robotics and Automation Letters (RA-L): Special Issue on Benchmarking Protocols for Robotic Manipulation.

Towards Robust Sensing for Autonomous Vehicles

Published in IEEE Transactions on Signal Processing Magazine (SPM): Special Issue on Autonomous Driving - To Appear, 2019, 2019

This paper surveys the emerging field of sensing in adversarial settings: after reviewing adversarial attacks on sensing modalities for autonomous systems, and discusses existing countermeasures and present future research directions.

Recommended citation: Apostolos Modas*, Ricardo Sanchez-Matilla*, Pascal Frossard, and Andrea Cavallaro. "Benchmark for Human-to-Robot Handovers of Unseen Containers with Unknown Filling." IEEE Transactions on Signal Processing Magazine (SPM): Special Issue on Autonomous Driving.

Multi-view shape estimation of transparent containers

Published in Submitted to IEEE Int. Conf. on Acoustics, Speech and Signal Processing (ICASSP), Barcelona, Spain, May 4-8, 2020

This paper proposes a method to estimate the shape and dimensions of unseen objects.

Recommended citation: A. Xompero, R. Sanchez-Matilla, A. Modas, P. Frossard, A. Cavallaro. "Multi-view shape estimation of transparent containers." Proc. of IEEE Int. Conf. on Acoustics, Speech and Signal Processing (ICASSP).

talks

Presentation at ACM Multimedia conference

Published:

Talk presenting the work titled Multi-modal localization and enhancement of multiple sound sources from a Micro Aerial Vehicle in the ACM Multimedia conference at Mountain View, USA.

Presentation at IROS conference

Published:

Talk presenting the work titled Tracking a moving sound source from a multi-rotor drone in the IROS conference at Madrid, Spain.

Presentation at ICIP conference

Published:

Talk presenting the work titled Confidence Intervals for Tracking Perforamance Scores in the ICIP conference at Athens, Greece.

Presentation at ICIP conference

Published:

Talk presenting the work titled A predictor of moving objects for first-person vision in the ICIP conference at Taipei, Taiwan.

Presentation at IROS conference

Published:

Talk presenting the work titled Audio-visual sensing from a quadcopter: dataset and baselines for source localization and sound enhancement in the IROS conference at Macau, China.

teaching

Teacher Assistant - Data Mining

Data Mining, Queen Mary University of London, 2018

Data that has relevance for decision-making is accumulating at an incredible rate due to a host of technological advances. Electronic data capture has become inexpensive and ubiquitous as a by-product of innovations such as the Internet, e-commerce, electronic banking, point-of-sale devices, bar-code readers, and electronic patient records. Data mining is a rapidly growing field that is concerned with developing techniques to assist decision-makers to make intelligent use of these repositories. The field of data mining has evolved from the disciplines of statistics and artificial intelligence. This module will combine practical exploration of data mining techniques with a exploration of algorithms, including their limitations. Students taking this module should have an elementary understanding of probability concepts and some experience of programming.

Teacher Assistant - Data Analytics

Data Analytics, Queen Mary University of London, 2019

This module focuses on the range of approaches, methodologies, techniques and tools for data analysis, and the use of data analysis findings to inform decision-making in an industrial / business context. It exposes students to a range of industry-standard statistical and data analysis techniques and tools, and fosters awareness of the challenges associated with working with large datasets. The module also covers topics related to the legal, social, ethical and professional issues associated with data storage and analysis. Students will undertake practical work including empirical data analysis and summarisation / presentation of the results to a range of relevant stakeholders.

Senior Teacher Assistant - Data Mining

Data Mining, Queen Mary University of London, 2019

Data that has relevance for decision-making is accumulating at an incredible rate due to a host of technological advances. Electronic data capture has become inexpensive and ubiquitous as a by-product of innovations such as the Internet, e-commerce, electronic banking, point-of-sale devices, bar-code readers, and electronic patient records. Data mining is a rapidly growing field that is concerned with developing techniques to assist decision-makers to make intelligent use of these repositories. The field of data mining has evolved from the disciplines of statistics and artificial intelligence. This module will combine practical exploration of data mining techniques with a exploration of algorithms, including their limitations. Students taking this module should have an elementary understanding of probability concepts and some experience of programming.