A linearization-based hybrid approach for 3D reconstruction of objects in a single image

Abstract

The shape-from-shading (SFS) technique uses the pattern of shading in images in order to obtain 3D view information. By virtue of their ease of implementation, linearization-based SFS algorithms are frequently used in the literature. In this study, Fourier coefficients of central differences obtained from gray-level images are employed, and two basic linearization-based algorithms are combined. By using the functionally generated surfaces and 3D reconstruction datasets, the hybrid algorithm is compared with linearization-based approaches. Five different evaluation metrics are applied on recovered depth maps and the corresponding gray-level images. The results on defective sample surfaces are also included to show the effect of the algorithm on surface reconstruction. The proposed method can prevent erroneous estimates on object boundaries and produce satisfactory 3D reconstruction results in a low number of iterations.

Accurate Analysis of the Spatial Pattern of Reflected Light and Surface Orientations Based on Color Illumination

“…Kotan et al [40] obtained the depth map by using different spatial coefficients of numerical gradients of images as an initial state and linearizing the reflectance map in terms of depth. A hybrid linearization based SFS is presented.…”

A comprehensive study of clustering a class of 2D shapes

“…The problem of shape and size analysis of 2D curves and functions is gaining increasing interest due to many important applications, such as cardiovascular analysis (Sangalli et al, 2010), finance interest rates (Kanevski and Timonin, 2010), nuclear industry (Auder and Fischer, 2012). Important to note are the advances in 3D surface reconstruction from images (Kotan et al, 2021).…”

Efficient face detection based crowd density estimation using convolutional neural networks and an improved sliding window strategy

“…The direct approaches (object/face detection based) attempt to segment and detect each person in crowd scenes and count them using effective detectors/classifiers (Zhao et al, 2008). Detecting people using this method becomes more difficult in a high-density crowd where occlusions occur (Kotan et al, 2021). Nevertheless, crowd counting under substantial occlusion problems has 8 R. Kian Ara et al been addressed adapting part-based detectors, such as a head detector (Sheng-Fuu et al, 2001) or a pedestrian detector (Khatoon et al, 2012).…”