Abstract
In this paper we present a model based video coder for remote video surveillance applications at very low bit rates. Unlike the known model based video codecs, we use the model based approach for background modeling and not for "head-and-shoulders" modeling. Since in video surveillance the remote site that is being monitored is usually a previously known physically closed space, we build a 3-D model of the site and use it to synthesize the background of the scene for each of the video cameras used. The codec has two layers, one layer for the background using a 3-D model and a second layer for the part of the scene not represented by the background. The second layer may use conventional hybrid DPCM-DCT coding schemes, like H.263, or object-based coding techniques, like those used in MPEG-4 video VM. The proposed codec can be a good framework for a multiview codec.

Abstract
Higher-order statistics (HOS) are a powerful analysis tool in digital signal processing. The most difficult task to use it effectively is the estimation of higher-order moments of sampled data, taken from real systems. For applications that require real-time processing, the performance achieved by common microprocessors or digital signal processors is not good enough to carry out the large number of calculations needed for their estimation. This paper presents ProHos-1, an experimental vector processor for the estimation of the higher-order moments up to the fourth-order. The processor's architecture exploits the structure of the algorithm, to process in parallel four vectors of the input data in a pipe-lined fashion, executing the equivalent to 11 operations in each clock cycle. The design of dedicated control circuits led to high clock rate and small hardware complexity, thus suitable for implementation as an ASIC (Application Specific integrated Circuit).

Abstract
The widespread use of smartphones and other low-cost equipment as recording devices, the massive growth in bandwidth, and the ever-growing demand for new applications with enhanced capabilities, made visual data a must in several scenarios, including surveillance, sports, retail, entertainment, and intelligent vehicles. Despite significant advances in analyzing and extracting data from images and video, there is a lack of solutions able to analyze and semantically describe the information in the visual scene so that it can be efficiently used and repurposed. Scientific contributions have focused on individual aspects or addressing specific problems and application areas, and no cross-domain solution is available to implement a complete system that enables information passing between cross-cutting algorithms. This paper analyses the problem from an end-to-end perspective, i.e., from the visual scene analysis to the representation of information in a virtual environment, including how the extracted data can be described and stored. A simple processing pipeline is introduced to set up a structure for discussing challenges and opportunities in different steps of the entire process, allowing to identify current gaps in the literature. The work reviews various technologies specifically from the perspective of their applicability to an end-to-end pipeline for scene analysis and synthesis, along with an extensive analysis of datasets for relevant tasks.

Abstract
The problem of gathering sufficiently representative data, such as those about human actions, shapes, and facial expressions, is costly and time-consuming and also requires training robust models. This has led to the creation of techniques such as transfer learning or data augmentation. However, these are often insufficient. To address this, we propose a semi-automated mechanism that allows the generation and editing of visual scenes with synthetic humans performing various actions, with features such as background modification and manual adjustments of the 3D avatars to allow users to create data with greater variability. We also propose an evaluation methodology for assessing the results obtained using our method, which is two-fold: (i) the usage of an action classifier on the output data resulting from the mechanism and (ii) the generation of masks of the avatars and the actors to compare them through segmentation. The avatars were robust to occlusion, and their actions were recognizable and accurate to their respective input actors. The results also showed that even though the action classifier concentrates on the pose and movement of the synthetic humans, it strongly depends on contextual information to precisely recognize the actions. Generating the avatars for complex activities also proved problematic for action recognition and the clean and precise formation of the masks.