Images, videos, audio signals, sensor data, can be easily collected in huge quantity by different devices and processed in order to emulate the human capability of elaborating a variety of different stimuli. Are multimodal signals useful to understand and anticipate human actions if acquired from the user viewpoint? This paper proposes to build an embedding space where inputs of different nature, but semantically correlated, are projected in a new representation space and properly exploited to anticipate the future user activity. 
To this purpose, we built a new multimodal dataset comprising video, audio, tri-axial acceleration, angular velocity, tri-axial magnetic field, pressure and temperature.
To benchmark the proposed multimodal anticipation challenge, we consider classic classifiers on top of deep learning methods used to build the embedding space representing multimodal signals.
The achieved results show that the exploitation of different modalities is useful to improve the anticipation of the future activity.

Anticipating Activity from Multimodal Signals

MMForWild2020 - MultiMedia FORensics in the WILD 2020

In our work, we focused on face identification from thermal images. Thermal face identification brings several advantages such as ease of handling with low resolution images, robust to illumination and privacy preserving.
Moreover, we considered scenarios in which face identification from thermal images is performed with low computation and memory resources, making such systems suitable for mobile or embedded applications such as STM32F7 family microcontroller.

Real-Time Thermal Face Identification System for Low Memory Vision Applications Using CNN

PS T3.3: Computer Vision 2D, 3D and Tracking

poster

ICPR 2020 is the premier international conference on Pattern Recognition, gathering leading researchers from around the world to present their latest achievements in Pattern Analysis and Recognition, applied Artificial Intelligence and Machine Learning. The topic of the year is " Putting Artificial intelligence to work on Patterns". 
For the first time, ICPR has employed a two-round review process similar to journal submissions, with new papers allowed to be submitted in either the first or second round and papers submitted in the first round and not accepted allowed to be revised and re-submitted for second round review. 

In this edition of the **ICPR 2020** Conference, there are 1407 Oral/Poster presentations organized in five Tracks: 
- Artificial Intelligence, Machine Learning for Pattern Analysis; 
- Biometrics, Human Analysis and Behavior Understanding; Computer Vision, Robotics and Intelligent Systems; 
- Document and Media Analysis; and 
- Image and Signal Processing. 

The **ICPR 2020** Main Conference program also includes 7 Keynotes from prominent world-renowned scientists, 8 Tutorials, 28 Demos, 8 Challenges, 2 Panels, as well as an Industry/Sponsor Exhibition. In addition, 41 Workshops on a range of important current and emerging research topics complete the ICPR 2020 program. 


ICPR 2020

[![](https://assets.underline.io/uploads/markdown_image/1/image/98bbd29b3a7633de61894a3a21feba0c.png)](https://iplab.dmi.unict.it/mmforwild/) [![](https://assets.underline.io/uploads/markdown_image/1/image/60d54d8c0e3dd52db4cf1815fa980546.png)](https://iplab.dmi.unict.it/mmforwild/programme) [![](https://assets.underline.io/uploads/markdown_image/1/image/0b3fc6f7e0ef64fdaa5dadaf5aa31479.png)](https://drive.google.com/file/d/14ryPDEosBw2L57ztnxFHPi-b2si0u7ey/view?usp=sharing)

The rapid growth of multimedia collections world-wide opens new challenges to multimedia technology with respect to crime prevention and investigation. The protection of images, video, and audio data from illegal use, as well as its exploitation in forensics and intelligence, have become serious challenges as the sheer data volume renders a full manual inspection by an expert impossible. Tools are needed to support the protection, management, processing, interpretation, and visualization of multimedia data in the different steps of the investigation process. Many exciting solutions for related problems have been developed in the multimedia research community (including knowledge extraction, categorization, indexing, browsing and visualization), forming an excellent basis for forensics and intelligence. The problem with many solutions developed so far is that they work well in controlled settings like those typical of laboratory experiments, but often fail to provide reliable answers in real-life conditions such as those encountered by forensic analysts and investigators in their daily activities.

MMForWILD offers a forum for proposing multimedia forensic solutions meeting the operational needs of forensics and intelligence operators. The workshop is targeted both at researchers working on innovative multimedia technology and experts developing tools in the field. The goal of the workshop is to attract and invite papers investigating the use of multimedia forensics outside the controlled environment of research laboratories (LEA, Practitioners, etc.). It intends to offer a venue for theory- and data-driven techniques addressing the trustworthiness of media data and the ability of verifying their integrity to prevent harmful misuses, seeking for solutions at the edge of signal processing, deep learning, multimedia analysis.

Topics of interests include, but are not limited to:

* Tampering Detection
* Source Identification
* Media Attribution
* Adversarial Multimedia forensics
* Fake vs Natural Imagery
* Digital Counter Forensics
* Audio Authentication
* Open set device attribution
* Digital forensics in the presence of incomplete information
* Detection of deep fake videos
* Interpretable forensics and Deep Learning
* Large scale forensic analysis
* Detection and identification of complex media processing chains
* Semantic-based forensic analysis

Sebastiano Battiato

3

SHORT BIO

Presentations

Anticipating Activity from Multimodal Signals

MMForWild2020 - MultiMedia FORensics in the WILD 2020

Real-Time Thermal Face Identification System for Low Memory Vision Applications Using CNN

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES