Yary Volpe

Improving surface reconstruction in shape from shading using easy-to-set boundary conditions

Minimisation techniques are commonly adopted methodologies for retrieving a 3D surface starting from its shaded representation (image), i.e., for solving the widely known shape from shading (SFS) problem. Unfortunately, depending on the imaged object to be reconstructed, retrieved surfaces often results to be completely different from the expected ones. In recent years, a number of interactive methods have been explored with the aim of improving surface reconstruction; however, since most of these methods require user interaction performed on a tentative reconstructed surface which often is significantly different from the desired one, it is advisable to increase the quality of the surface, to be further processed, as much as possible. Inspired by such techniques, the present work describes a new method for interactive retrieving of shaded object surface. The proposed approach is meant to recover the expected surface by using easy-to-set boundary conditions, so that the human-computer interaction primaril...

From 2D to 2.5D i.e. from painting to tactile model

Display Omitted Commonly used to produce the visual effect of full 3D scene on reduced depth supports, bas relief can be successfully employed to help blind people to access inherently bi-dimensional works of art. Despite a number of methods have been proposed dealing with the issue of recovering 3D or 2.5D surfaces from single images, only a few of them explicitly address the recovery problem from paintings and, more specifically, the needs of visually impaired and blind people.The main aim of the present paper is to provide a systematic method for the semi-automatic generation of 2.5D models from paintings. Consequently, a number of ad hoc procedures are used to solve most of the typical problems arising when dealing with artistic representation of a scene. Feedbacks provided by a panel of end-users demonstrated the effectiveness of the method in providing models reproducing, using a tactile language, works of art otherwise completely inaccessible.

Digital Bas-Relief Design: a Novel Shape from Shading-Based Method

Design of products characterized by high stylistic content and organic shapes in the form of bas-relief (e.g. fashion accessories, commemorative plaques and coins) is traditionally performed starting from handmade drawings or photographs that are manually reproduced by highly skilled craftsmen such as sculptors and engravers and finally digitized by means of 3D scanning. Several Computer-based procedures have been devised with the aim of speeding up this process, which is considerably time consuming, subjective and costly; these are mainly based on image processing techniques such as embossing, enhancement, histogram equalization or dynamic range, also implemented in CAD-based commercial software. However, these approaches are characterized by several limitations preventing them from providing a “correct” final geometry. In view of that, the present work describes a novel method for the creation of digital bas-reliefs from a single image using a Shape From Shading (SFS) based approach with interactive ini...

Tactile representation of paintings: an early assessment of possible computer based strategies

In recent years, a number of works meant to define the criteria for translating two-dimensional art into tactile representation, to be benefit of blind and visually impaired people. Due to technology-related limitations, however, these studies mainly investigated only some kinds of possible representations (e.g. tactile diagrams). This work deals with the analysis of 4 alternative translation strategies, implemented using computer-based tools, to determine the most effective one in delivering blind people a correct perception of pictorial artworks. The outcomes of the study contribute new information to the field of tactile paintings for blind and visually impaired individuals by testing the response of a panel of potential users.

Modelling and simulation of an innovative fabric coating process using artificial neural networks

A polyurethane-based fabric coating process requires a series of parameters to be set in order to meet the desired quality of the final product. Usually, the optimal setting of such parameters is performed by means of experimental tests, based on the experience of trained operators. The lack of understanding of the interaction between the coating process parameters and the final quality properties of the coated fabric encourages the development of predictive models. The main aim of the present work is to provide a predictive model of a particular coating process for forecasting the final characteristics of a coated fabric, based on the process parameters. The devised model, based on artificial neural networks, is trained and validated using a wide experimental database created with reference to an innovative coating process. Once simulated with new process parameters, the model proves to be capable of determining the best possible process parameter values to obtain the preferred coated fabric properties. By employing the developed model, a series of charts are also built that can be used to provide technicians with a practical tool for effectively selecting the process parameters.

A RGB-D based instant body-scanning solution for compact box installation

Body scanning presents unique value in delivering the first digital asset of a human body thus resulting a fundamental device for a range of applications dealing with health, fashion and fitness. Despite several body scanners are in the market, recently depth cameras such as Microsoft Kinect® have attracted the 3D community; compared with conventional 3D scanning systems, these sensors are able to capture depth and RGB data at video rate and even if quality and depth resolution are not optimal for this kind of applications, the major benefit comes from the overall acquisition speed and from the IR pattern that allows poor lighting conditions optimal acquisition. When dealing with non-rigid bodies, unfortunately, the use of a single depth camera may lead to inconsistent results mainly caused by wrong surfaces registration. With the aim of improving existing systems based on low-resolution depth cameras, the present paper describes a novel scanning system for capturing 3D full human body models by using multiple Kinect® devices in a compact setup. The system consists of an instantaneous scanning system using eight depth cameras, appropriately arranged in a compact wireframe. To validate the effectiveness of the proposed architecture, a comparison of the obtained 3D body model with the one obtained using a professional Konica Minolta Range Seven 3D scanner is also presented and possible drawbacks are hinted at.

On the Performance of the Intel SR30 Depth Camera: Metrological and Critical Characterization

Specifically conceived for applications related to face analytics and tracking, scene segmentation, hand/finger tracking, gaming, augmented reality, and RGB-D cameras are nowadays used even as 3-D scanners. Despite depth cameras’ accuracy and precision are not comparable with professional 3-D scanners, they still constitute a promising device for reverse engineering (RE) applications in the close range, due to their low cost. This is particularly true for more recent devices, such as, for instance, the RealSense SR300, which promises to be among the best performing close range depth cameras in the market. Given the potentiality of this new device, and since to date a deep investigation on its performances has not been assessed in scientific literature, the main aim of this paper is to characterize and to provide metrological considerations on the Intel RealSense SR300 depth sensor when this is used as a 3-D scanner. To this end, the device sensor performances are first assessed by applying the existing normative guidelines (i.e. the one published by the Association of German Engineers - Verein Deutscher Ingenieure - VDI/VDE 2634) both to a set of raw captured depth data and to a set acquired with optimized setting of the camera. Then, further assessment of the device performances is carried out by applying some strategies proposed in the literature using optimized sensor setting, to reproduce “real life” conditions for the use as a 3-D scanner. Finally, the performance of the device is critically compared against the performance of latest short-range sensors, thus providing a useful guide, for researchers and practitioners, in an informed choice of the optimal device for their own RE application.

Computer-based methodologies for semi-automatic 3D model generation from paintings

Over the last few years, technologies like 3D scanning and rapid prototyping provided an extraordinary boost in improving reproductions of 3D artworks, like sculptures and historical buildings, all over the world. Physical 3D reproduction of subjects represented in paintings, is recognised to be one of the best ways to allow visually impaired people to enjoy such kind of artworks. However, the use of advanced technologies with the aim of realising 3D models starting from paintings has not been satisfactorily investigated yet. Though a number of algorithms coming from computer vision science exist to cope with similar issues, the specific problem of producing a 3D representation which is targeted at blind people tactile exploration has been only marginally investigated. Starting from these considerations, this work presents 1) a quite extensive review of the criteria proposed in literature for producing tactile models suitable for blind people and 2) four alternative computer-based methods for semi-automatic generation of tactile 3D models starting from RGB digital images of paintings. The outcomes of this study contribute new information to the field of visually impaired user-oriented 3D reconstruction and clearly indicate the strategy to be adopted in order to produce a meaningful reproduction of a bi-dimensional piece of artwork.

Methods for Predicting Spectral Response of Fibers Blends

Textile companies usually manufacture fabrics using a mix of pre-colored fibers according to a traditional recipe based on their own experience. Unfortunately, mainly due to the fibers dyeing process, the colorimetric distance between the obtained fabric and the desired one results unsatisfactory with respect to a colorimetric threshold established by the technicians. In such cases, colorists are required to slightly change the original recipe in order to reduce the colorimetric distance. This trial and error process is time-consuming and requires the work of highly skilled operators. Computer-based color recipe assessment methods have been proposed so far in scientific literature to address this issue. Unlikely, many methods are still far to be reliably predictive when the fabric is composed by a high number of components. Accordingly, the present work proposes two alternative methods based on Kubelka-Munk and subtractive mixing able to perform a reliable prediction of the spectrophotometric response of a fabric obtained by means of any variation of a recipe. The assessment performed on a prototypal implementation of the two methods demonstrates that they are suitable for reliable prediction of fabric blends spectral response.

Color matching of fabric blends: hybrid Kubelka-Munk + artificial neural network based method

Abstract. Color matching of fabric blends is a key issue for the textile industry, mainly due to the rising need to create high-quality products for the fashion market. The process of mixing together differently colored fibers to match a desired color is usually performed by using some historical recipes, skillfully managed by company colorists. More often than desired, the first attempt in creating a blend is not satisfactory, thus requiring the experts to spend efforts in changing the recipe with a trial-and-error process. To confront this issue, a number of computer-based methods have been proposed in the last decades, roughly classified into theoretical and artificial neural network (ANN)–based approaches. Inspired by the above literature, the present paper provides a method for accurate estimation of spectrophotometric response of a textile blend composed of differently colored fibers made of different materials. In particular, the performance of the Kubelka-Munk (K-M) theory is enhanced by introducing an artificial intelligence approach to determine a more consistent value of the nonlinear function relationship between the blend and its components. Therefore, a hybrid K-M+ANN-based method capable of modeling the color mixing mechanism is devised to predict the reflectance values of a blend.