Publications

Most of the public satellite image datasets contain only a small number of annotated images. The lack of a sufficient quantity of labeled data for training is a bottleneck for the use of modern deep-learning based classification approaches in this domain. In this paper we propose a semi -supervised approach to deal with this problem. We use the discriminator $(D)$ of a Generative Adversarial Network (GAN) as the final classifier, and we train $D$ using both labeled and unlabeled data. The main novelty we introduce is the representation of the visual information fed to $D$ by means of two different channels: the original image and its “semantic” representation, the latter being obtained by means of an external network trained on ImageNet. The two channels are fused in $D$ and jointly used to classify fake images, real labeled and real unlabeled images. We show that using only 100 labeled images, the proposed approach achieves an accuracy close to 69% and a significant improvement with respect to other GAN-based semi-supervised methods. Although we have tested our approach only on satellite images, we do not use any domain-specific knowledge. Thus, our method can be applied to other semi-supervised domains.

Vehicle re-identification plays a major role in modern smart surveillance systems. Specifically, the task requires the capability to predict the identity of a given vehicle, given a dataset of known associations, collected from different views and surveillance cameras. Generally, it can be cast as a ranking problem: given a probe image of a vehicle, the model needs to rank all database images based on their similarities w.r.t the probe image. In line with recent research, we devise a metric learning model that employs a supervision based on local constraints. In particular, we leverage pairwise and triplet constraints for training a network capable of assigning a high degree of similarity to samples sharing the same identity, while keeping different identities distant in feature space. Eventually, we show how vehicle tracking can be exploited to automatically generate a weakly labelled dataset that can be used to train the deep network for the task of vehicle re-identification. Learning and evaluation is carried out on the NVIDIA AI city challenge videos.

A long tradition in social psychology focused on nonverbal behaviour displayed during dyadic interactions generally relying on evaluations from external coders. However, in addition to the fact that external coders may be biased, they may not capture certain type of behavioural indices. We designed three studies examining explicit and implicit prejudice as predictors of nonberval behaviour as reflected in objective indices provided by Kinect cameras. In the first study, we considered White-Black relations from the perspective of 36 White participants. Results revealed that implicit prejudice was associated with a reduction in interpersonal distance and in the volume of space between Whites and Blacks (vs. Whites and Whites), which in turn were associated with evaluations by collaborators taking part in the interaction. In the second study, 37 non-HIV participants interacted with HIV individuals. We found that implicit prejudice was associated with reduced volume of space between interactants over time (a process of bias overcorrection) only when they tried hard to control their behaviour (as captured by a stroop test). In the third study 35 non-disabled children interacted with disabled children. Results revealed that implicit prejudice was associated with reduced interpersonal distance over time.

Dematerialization and digitalization of historical documents are key elements for their availability, preservation and diffusion. Unfortunately, the conversion from handwritten to digitalized documents presents several technical challenges. The XDOCS project is created with the main goal of making available and extending the usability of historical documents for a great variety of audience, like scholars, institutions and libraries. In this paper the core elements of XDOCS, i.e. page dewarping and word spotting technique, are described and two new applications, i.e. annotation/indexing and search tool, are presented.

US-guided neuronavigation exploits the simplicity of use and minimal invasiveness of Ultrasound (US) imaging and the high tissue resolution and signal-to-noise ratio of Magnetic Resonance Imaging (MRI) to guide brain surgeries. More specifically, the intra-operative 3D US images are combined with pre-operative MR images to accurately localise the course of instruments in the operative field with minimal invasiveness. Multi-modal image registration of 3D US and MR images is an essential part of such system. In this paper, we present a complete software framework that enables the registration US and MR brain scans based on a multi resolution deformable transform, tackling elastic deformations (i.e. brain shifts) possibly occurring during the surgical procedure. The framework supports also simpler and faster registration techniques, based on rigid or affine transforms, and enables the interactive visualisation and rendering of the overlaid US and MRI volumes. The registration was experimentally validated on a public dataset of realistic brain phantom images, at different levels of artificially induced deformations.

We examined predictors and consequences of intergroup nonverbal behaviour by relying on new technologies and new objective indices. In three studies, both in the laboratory and in the field with children, behaviour was a function of implicit prejudice.

We discuss modelling issues related to the design of a somatic facial motor space. The variants proposed are conceived to be part of a larger system for dealing with simulation-based face emotion analysis along dual interactions.

We present a novel video browsing and retrieval system for edited videos, in which videos are automatically decomposed into meaningful and storytelling parts (i.e. scenes) and tagged according to their transcript. The system relies on a Triplet Deep Neural Network which exploits multimodal features, and has been implemented as a set of extensions to the eXo Platform Enterprise Content Management System (ECMS). This set of extensions enable the interactive visualization of a video, its automatic and semi-automatic annotation, as well as a keyword-based search inside the video collection. The platform also allows a natural integration with third-party add-ons, so that automatic annotations can be exploited outside the proposed platform.