[Paper Review] Wavelets and continuous wavelet transform for autostereoscopic multiview images

Vladimir Saveljev|arXiv (Cornell University)|Jun 8, 2015

Advanced Optical Imaging Technologies17 references3 citations

TL;DR

This paper proposes a wavelet-based approach using continuous wavelet transform (CWT) to analyze autostereoscopic multiview and integral images, leveraging reference functions as scaling functions for wavelet construction. The method successfully detects structural features in binary wireframe test objects, with restored locations closely matching the original object geometry, demonstrating potential for 3D image analysis in multiview displays.

ABSTRACT

Recently, the reference functions for the synthesis and analysis of the autostereoscopic multiview and integral images in three-dimensional displays we introduced. In the current paper, we propose the wavelets to analyze such images. The wavelets are built on the reference functions as on the scaling functions of the wavelet analysis. The continuous wavelet transform was successfully applied to the testing wireframe binary objects. The restored locations correspond to the structure of the testing wireframe binary objects.

Motivation & Objective

• To develop a wavelet-based analysis framework for autostereoscopic multiview and integral images.
• To utilize reference functions from prior work as scaling functions in wavelet construction.
• To apply continuous wavelet transform (CWT) for detecting structural features in 3D image data.
• To validate the method on binary wireframe test objects to assess accuracy in feature localization.
• To demonstrate the feasibility of wavelet transforms in analyzing 3D display image content.

Proposed method

• Wavelets are constructed using reference functions as scaling functions, forming the basis for multiresolution analysis.
• The continuous wavelet transform (CWT) is applied to test images to detect localized features across scales and positions.
• The analysis focuses on binary wireframe objects to evaluate structural detection performance.
• Feature localization is determined by identifying maxima in the CWT coefficient magnitude across scales.
• The method relies on the geometric consistency between detected features and original object structures.
• The approach is evaluated using visual inspection and comparison of restored locations with ground truth.

Experimental results

Research questions

• RQ1 Can wavelets built from reference functions effectively analyze autostereoscopic multiview images?
• RQ2 How accurately does the continuous wavelet transform detect structural features in binary wireframe objects?
• RQ3 To what extent do the restored feature locations correspond to the actual geometry of the test objects?
• RQ4 Can the wavelet-based method preserve spatial relationships in 3D image data?
• RQ5 Is the CWT approach suitable for feature extraction in integral and multiview 3D image processing?

Key findings

• The continuous wavelet transform successfully detected structural features in binary wireframe test objects.
• Restored feature locations closely matched the actual positions in the original wireframe objects.
• The method demonstrated sensitivity to geometric details through scale-space analysis.
• The use of reference functions as scaling functions enabled effective wavelet construction for 3D image analysis.
• The results indicate that wavelet transforms can be a viable tool for analyzing multiview and integral images.
• The approach shows promise for applications in 3D display technology and image feature extraction.