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

Abstract

a b s t r a c tCommonly 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 semiautomatic 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.

Computer‐assisted Relief Modelling: A Comprehensive Survey

“…Using frequency analysis of image intensity to convey shape information suffers from the concave‐convex ambiguity. To overcome this problem, Furferi et al [FGV∗14] present a systematic method combining perspective geometry‐based scene reconstruction and SFS‐based volume reconstruction to generate 2.5D tactile models. After image segmentation, the height map of all subjects in the painting is obtained as a combination of three different height maps: rough shape, main shape and fine details shape.…”

“…Region‐based approach operates at region level, and has the advantage of wider applications with various input forms, e.g. photographs [YHJ∗16], oil paintings [RMP11] [FGV∗14], hand‐drawn cartoons [SKJ∗14] [DNJ∗18], rubbing images [LWYM12], brush paintings [FYY∗18] and line drawings [KLST11]. Another advantage of region‐based approach is its explicit consideration of the regions depth ordering, which generates reliefs with more plausible overall shapes and easier for 3D printing.…”

“…However, inflation alone cannot recover the fine details in shape, which is a limitation of pure region‐based methods. In some works, fine details are produced by considering the pixel brightness [LWYM12] [RMP11] or the gradient magnitude [ZMW∗13] [FGV∗14] of the input image as a height map, and or using SFS to generate a displacement map [FYY∗18] [FGV∗14], which will be combined with the inflation surface afterwards to generate a textured relief. Table summarizes characteristics of image‐space relief modelling methods.…”

“…The other is the manual depth ordering, which aims to construct a reasonable height relations among neighboring regions. This is achieved usually by user interventions such as depth inequalities definition [SKJ∗14], T‐junctions and overlap analysis [YHJ∗16], directed acyclic graph marking [RMP11], region classification [FGV∗14] or edge directing [KLST11]. During surface inflation, user markup cues are also necessary so as to generate the correct concave‐convex profile of the surface.…”

Ubiquitous Creation of Bas-Relief Surfaces with Depth-of-Field Effects Using Smartphones

“…For this reason, algorithms have been designed for bas-relief generation based upon inputs in the form of images. The specific types of image inputs can include rubbing images generated by brick and stone relief [10] or painting images [11,12]. However, the most common form of the input image is a photograph, which can be obtained anywhere by using a smartphone.…”