Abstract
In this paper, a method for estimation of human pose is proposed, making use of ToF (Time of Flight) cameras. For this, a YOLO based object detection method was used, to develop a top-down method. In the first stage, a network was developed to detect people in the image. In the second stage, a network was developed to estimate the joints of each person, using the image result from the first stage. We show that a deep learning network trained from scratch with ToF images yields better results than taking a deep neural network pretrained on RGB data and retraining it with ToF data. We also show that a top-down detector, with a person detector and a joint detector works better than detecting the body joints over the entire image.

Abstract
The fusion of pre-operative 3D magnetic resonance (MR) images with real-time 3D ultrasound (US) images can be the most beneficial way to guide minimally invasive cardiovascular interventions without radiation. Previously, we addressed this topic through a strategy to segment the left ventricle (LV) on interventional 3D US data using a personalized shape prior obtained from a pre-operative MR scan. Nevertheless, this approach was semi-automatic, requiring a manual alignment between US and MR image coordinate systems. In this paper, we present a novel solution to automate the abovementioned pipeline. In this sense, a method to automatically detect the right ventricular (RV) insertion point on the US data was developed, which is subsequently combined with pre-operative annotations of the RV position in the MR volume, therefore allowing an automatic alignment of their coordinate systems. Moreover, a novel strategy to ensure a correct temporal synchronization of the US and MR models is applied. Finally, a full evaluation of the proposed automatic pipeline is performed. The proposed automatic framework was tested in a clinical database with 24 patients containing both MR and US scans. A similar performance between the proposed and the previous semi-automatic version was found in terms of relevant clinical measurements. Additionally, the automatic strategy to detect the RV insertion point showed its effectiveness, with a good agreement against manually identified landmarks. Overall, the proposed automatic method showed high feasibility and a performance similar to the semi-automatic version, reinforcing its potential for normal clinical routine.

Abstract
Deformational Plagiocephaly (DP) refers to an asymmetrical distortion of an infant's skull resulting from external forces applied over time. The diagnosis of this condition is performed using asymmetry indexes that are estimated from specific anatomical landmarks, whose are manually defined on head models acquired using laser scans. However, this manual identification is susceptible to intra-/inter-observer variability, being also time-consuming. Therefore, automatic strategies for the identification of the landmarks and, consequently, extraction of asymmetry indexes, are claimed. A novel pipeline to automatically identify these landmarks on 3D head models and to estimate the relevant cranial asymmetry indexes is proposed. Thus, a template database is created and then aligned with the unlabelled patient through an iterative closest point (ICP) strategy. Here, an initial rigid alignment followed by an affine one are applied to remove global misalignments between each template and the patient. Next, a non-rigid alignment is used to deform the template information to the patient-specific shape. The final position of each landmark is computed as a local weight average of all candidate results. From the identified landmarks, a head's coordinate system is automatically estimated and later used to estimate cranial asymmetry indexes. The proposed framework was evaluated in 15 synthetic infant head's model. Overall, the results demonstrated the accuracy of the identification strategy, with a mean average distance of 2.8 +/- 0.6 mm between the identified landmarks and the ground-truth. Moreover, for the estimation of cranial asymmetry indexes, a performance comparable to the inter-observer variability was achieved.

Abstract
The overall objective of this work is to review the assistive technologies that have been proposed by researchers in recent years to address the limitations in user mobility posed by visual impairment. This work presents an umbrella review. Visually impaired people often want more than just information about their location and often need to relate their current location to the features existing in the surrounding environment. Extensive research has been dedicated into building assistive systems. Assistive systems for human navigation, in general, aim to allow their users to safely and efficiently navigate in unfamiliar environments by dynamically planning the path based on the user's location, respecting the constraints posed by their special needs. Modern mobile assistive technologies are becoming more discrete and include a wide range of mobile computerized devices, including ubiquitous technologies such as mobile phones. Technology can be used to determine the user's location, his relation to the surroundings (context), generate navigation instructions and deliver all this information to the blind user.

Abstract
Information can be conveyed to the user by means of a narrative, modeled according to the user’s context. A case in point is the weather, which can be perceived differently and with distinct levels of importance according to the user’s context. For example, for a blind person, the weather is an important element to plan and move between locations. In fact, weather can make it very difficult or even impossible for a blind person to successfully negotiate a path and navigate from one place to another. To provide proper information, narrated and delivered according to the person’s context, this paper proposes a project for the creation of weather narratives, targeted at specific types of users and contexts. The proposal’s main objective is to add value to the data, acquired through the observation of weather systems, by interpreting that data, in order to identify relevant information and automatically create narratives, in a conversational way or with machine metadata language. These narratives should communicate specific aspects of the evolution of the weather systems in an efficient way, providing knowledge and insight in specific contexts and for specific purposes. Currently, there are several language generator’ systems, which automatically create weather forecast reports, based on previously processed and synthesized information. This paper, proposes a wider and more comprehensive approach to the weather systems phenomena, proposing a full process, from the raw data to a contextualized narration, thus providing a methodology and a tool that might be used for various contexts and weather systems.

Abstract
This study has the objective of select the best service at image processing and recognition, running in the cloud, and best suited for usage in systems to aid and improve the daily lives of blind people. To accomplish this purpose, a set of candidate services was built, including Microsoft Cognitive Services and Google Cloud Vision. A test mobile app was developed to automatically take pictures, which are sent to the online cloud services for processing. The results and the functionalities were evaluated with the aim to measure their accuracy and relevance. The following variables were registered: relative accuracy, represented by the ratio of the number of accurate results vs. the number of results shown; confidence degree, representing the service accuracy (when provided by the service); and relevance, identifying situations that can be useful in the daily lives of the blind people. The results have shown that these two services, Microsoft Cognitive Services and Google Cloud Vision, provided good accuracy and significance, in supporting systems to help blind people in their daily tasks. It was chosen some functionalities in two APIs of services running in the cloud like face identification, image description, objects, and text recognition. © BEIESP.