Publications

In this paper we present a segmentation proposal method which employs a box-hypotheses generation step followed by a lightweight segmentation strategy. Inspired by interactive segmentation, for each automatically placed bounding-box we compute a precise segmentation mask. We introduce diversity in segmentation strategies enhancing a generic model performance exploiting class-independent regional appearance features. Foreground probability scores are learned from groups of objects with peculiar characteristics to specialize segmentation models. We demonstrate results comparable to the state-of-the-art on PASCAL VOC 2012 and a further improvement by merging our proposals with those of a recent solution. The ability to generalize to unseen object categories is demonstrated on Microsoft COCO 2014.

This work presents a research project, named XDOCS, aimed at extending to a much wider audience the possibility to access a variety of historical documents published on the web. The paper presents an overview of the indexing process that will be used to achieve the goal, focusing on the adopted dewarping technique. The proposed dewarping approach performs its task with the help of a transformation model which maps the projection of a curved surface to a 2D rectangular area. The novelty introduced with this work regards the possibility of applying dewarping to document images which contain both handwritten and typewritten text.

In this paper we address the abnormality detection problem in crowded scenes. We propose to use Generative Adversarial Nets (GANs), which are trained using normal frames and corresponding optical-flow images in order to learn an internal representation of the scene normality. Since our GANs are trained with only normal data, they are not able to generate abnormal events. At testing time the real data are compared with both the appearance and the motion representations reconstructed by our GANs and abnormal areas are detected by computing local differences. Experimental results on challenging abnormality detection datasets show the superiority of the proposed method compared to the state of the art in both frame-level and pixel-level abnormality detection tasks.

With the spread of wearable devices and head mounted cameras, a wide range of application requiring precise user localization is now possible. In this paper we propose to treat the problem of obtaining the user position with respect to a known environment as a video registration problem. Video registration, i.e. the task of aligning an input video sequence to a pre-built 3D model, relies on a matching process of local keypoints extracted on the query sequence to a 3D point cloud. The overall registration performance is strictly tied to the actual quality of this 2D-3D matching, and can degrade if environmental conditions such as steep changes in lighting like the ones between day and night occur. To effectively register an egocentric video sequence under these conditions, we propose to tackle the source of the problem: the matching process. To overcome the shortcomings of standard matching techniques, we introduce a novel embedding space that allows us to obtain robust matches by jointly taking into account local descriptors, their spatial arrangement and their temporal robustness. The proposal is evaluated using unconstrained egocentric video sequences both in terms of matching quality and resulting registration performance using different 3D models of historical landmarks. The results show that the proposed method can outperform state of the art registration algorithms, in particular when dealing with the challenges of night and day sequences.

Image and video captioning are important tasks in visual data analytics, as they concern the capability of describing visual content in natural language. They are the pillars of query answering systems, improve indexing and search and allow a natural form of human-machine interaction. Even though promising deep learning strategies are becoming popular, the heterogeneity of large image archives makes this task still far from being solved. In this paper we explore how visual saliency prediction can support image captioning. Recently, some forms of unsupervised machine attention mechanisms have been spreading, but the role of human attention prediction has never been examined extensively for captioning. We propose a machine attention model driven by saliency prediction to provide captions in images, which can be exploited for many services on cloud and on multimedia data. Experimental evaluations are conducted on the SALICON dataset, which provides groundtruths for both saliency and captioning, and on the large Microsoft COCO dataset, the most widely used for image captioning.

In the last years video has been swamping the Internet: websites, social networks, and business multimedia systems are adopting video as the most important form of communication and information. Video are normally accessed as a whole and are not indexed in the visual content. Thus, they are often uploaded as short, manually cut clips with user-provided annotations, keywords and tags for retrieval. In this paper, we propose a prototype multimedia system which addresses these two limitations: it overcomes the need of human intervention in the video setting, thanks to fully deep learning-based solutions, and decomposes the storytelling structure of the video into coherent parts. These parts can be shots, key-frames, scenes and semantically related stories, and are exploited to provide an automatic annotation of the visual content, so that parts of video can be easily retrieved. This also allows a principled re-use of the video itself: users of the platform can indeed produce new storytelling by means of multi-modal presentations, add text and other media, and propose a different visual organization of the content. We present the overall solution, and some experiments on the re-use capability of our platform in edutainment by conducting an extensive user valuation %with students from primary schools.

In this paper, we aim to understand whether current language and vision (LaVi) models truly grasp the interaction between the two modalities. To this end, we propose an extension of the MS-COCO dataset, FOIL-COCO, which associates images with both correct and ‘foil’ captions, that is, descriptions of the image that are highly similar to the original ones, but contain one single mistake (‘foil word’). We show that current LaVi models fall into the traps of this data and perform badly on three tasks: a) caption classification (correct vs. foil); b) foil word detection; c) foil word correction. Humans, in contrast, have near-perfect performance on those tasks. We demonstrate that merely utilising language cues is not enough to model FOIL-COCO and that it challenges the state-of-the-art by requiring a fine-grained understanding of the relation between text and image.

Recently, social theories and empirical observations identified small groups and leaders as the basic elements which shape a crowd. This leads to an intermediate level of abstraction that is placed between the crowd as a flow of people, and the crowd as a collection of individuals. Consequently, automatic analysis of crowds in computer vision is also experiencing a shift in focus from individuals to groups and from small groups to their leaders. In this chapter, we present state-of-the-art solutions to the groups and leaders detection problem, which are able to account for physical factors as well as for sociological evidence observed over short time windows. The presented algorithms are framed as structured learning problems over the set of individual trajectories. However, the way trajectories are exploited to predict the structure of the crowd is not fixed but rather learned from recorded and annotated data, enabling the method to adapt these concepts to different scenarios, densities, cultures, and other unobservable complexities. Additionally, we investigate the relation between leaders and their groups and propose the first attempt to exploit leadership as prior knowledge for group detection.