Three-dimensional range geometry compression via reduced entropy encoding of the image

Area of Science:

• Computer Vision
• Data Compression
• Geometric Modeling

Background:

• Three-dimensional (3D) data compression is crucial for efficient storage and transmission.
• Existing methods embed 3D range geometry into 2D images, improving compression ratios.
• Maximizing image capacity (color channels, bit depth) and using advanced algorithms like Free Lossless Image Format (FLIF) are current focuses.

Purpose of the Study:

• To propose a new approach for 3D data compression that prioritizes understanding image data structure and redundancy.
• To demonstrate the superiority of this method over conventional compression algorithms.
• To achieve significant improvements in lossless compression for 3D data.

Main Methods:

• Analyzing the inherent structure and redundancy within image data.
• Modifying the process of 2D image creation and compression.
• Exploiting reduced information entropy for enhanced compression.

Main Results:

• A significant reduction in the entropy of image information was achieved.
• The proposed method demonstrates robustness and effectiveness in lossless compression.
• An impressive compression ratio of 0.29 bit per point was obtained for objects with continuous smooth surfaces.

Conclusions:

• Modifying the 2D image creation and compression process by analyzing data structure offers superior lossless compression.
• This approach unlocks new opportunities for improving 3D data compression efficiency.
• The experimental results validate the proposed method's effectiveness and robustness.