Two-channel depth encoding for 3D range geometry compression

Finley, Matthew G.; Bell, Tyler

doi:10.1364/ao.58.006882

Cited by 9 publications

(7 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although the MWD method and similar algorithms are able to achieve high compression ratios when compared to the original floating-point 3D-range data or its mesh representation, several approaches have been proposed that attempt to reduce the number of encoding signals required to faithfully decode the 3D information from the color channels of a 2D RGB image [11][12][13][14][15]. These methods are able to decode compressed depth information from only two of the three color channels available within a traditional image format through either the removal of redundant encoded depth information or through the removal of the auxiliary information required by typical phase unwrapping algorithms.…”

Section: Image-based Range Geometry Compressionmentioning

confidence: 99%

“…The method proposed in this section enables the recovery of depth information from a single-channel depth encoding using a semantic segmentation approach. This is accomplished by first calculating the unsigned wrapped phase from a single sinusoidal encoding in the same fashion as that demonstrated in [14]. This process can be described for each pixel (i, j) as…”

Section: Single-channel Depth Decoding With Semantic Segmentationmentioning

confidence: 99%

“…Figure 2g is the wrapped phase, φ, where the sign of each pixel is determined using γ (Figure 2f) according to Equation (11). Figure 2h is a 3D rendering of Z , the depth information recovered from φ using γ according to Equations ( 12)- (14).…”

Section: Single-channel Depth Decoding With Semantic Segmentationmentioning

confidence: 99%

“…The third color channel is then used to store information required to recover the original 3D data from the two encoded signals. Additional methods have been proposed that aim to reduce file sizes by encoding 3D information in a manner that only requires two of the three color channels available in an RGB image to store and recover 3D-range data [11][12][13][14][15].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

SCDeep: Single-Channel Depth Encoding for 3D-Range Geometry Compression Utilizing Deep-Learning Techniques

et al. 2022

Self Cite

View full text Add to dashboard Cite

Recent advances in optics and computing technologies have encouraged many applications to adopt the use of three-dimensional (3D) data for the measurement and visualization of the world around us. Modern 3D-range scanning systems have become much faster than real-time and are able to capture data with incredible precision. However, increasingly fast acquisition speeds and high fidelity data come with increased storage and transmission costs. In order to enable applications that wish to utilize these technologies, efforts must be made to compress the raw data into more manageable formats. One common approach to compressing 3D-range geometry is to encode its depth information within the three color channels of a traditional 24-bit RGB image. To further reduce file sizes, this paper evaluates two novel approaches to the recovery of floating-point 3D range data from only a single-channel 8-bit image using machine learning techniques. Specifically, the recovery of depth data from a single channel is enabled through the use of both semantic image segmentation and end-to-end depth synthesis. These two distinct approaches show that machine learning techniques can be utilized to enable significant file size reduction while maintaining reconstruction accuracy suitable for many applications. For example, a complex set of depth data encoded using the proposed method, stored in the JPG 20 format, and recovered using semantic segmentation techniques was able to achieve an average RMS reconstruction accuracy of 99.18% while achieving an average compression ratio of 106:1 when compared to the raw floating-point data. When end-to-end synthesis techniques were applied to the same encoded dataset, an average reconstruction accuracy of 99.59% was experimentally demonstrated for the same average compression ratio.

show abstract

Section: Image-based Range Geometry Compressionmentioning

confidence: 99%

Section: Single-channel Depth Decoding With Semantic Segmentationmentioning

confidence: 99%

Section: Single-channel Depth Decoding With Semantic Segmentationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

SCDeep: Single-Channel Depth Encoding for 3D-Range Geometry Compression Utilizing Deep-Learning Techniques

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…Another approach to compressing 3D range data is to create 2D encodings based directly on the depth map, Z. Methods that follow this approach [16,17,18,19] directly encode the 3D range data's floatingpoint depth map using principles of phase-shifting, such that one or more high-frequency 8-bit encodings are produced. These encodings are then stored in one or two of the output 2D image's color channels.…”

Section: Introductionmentioning

confidence: 99%

Depth Range Reduction for 3D Range Geometry Compression

Finley,

Bell

2020

Preprint

Self Cite

View full text Add to dashboard Cite

Three-dimensional (3D) shape measurement devices and techniques are being rapidly adopted within a variety of industries and applications. As acquiring 3D range data becomes faster and more accurate it becomes more challenging to efficiently store, transmit, or stream this data. One prevailing approach to compressing 3D range data is to encode it within the color channels of regular 2D images. This paper presents a novel method for reducing the depth range of a 3D geometry such that it can be stored within a 2D image using lower encoding frequencies (or a fewer number of encoding periods). This allows for smaller compressed file sizes to be achieved without a proportional increase in reconstruction errors. Further, as the proposed method occurs prior to encoding, it is readily compatible with a variety of existing image-based 3D range geometry compression methods.

show abstract

Depth range reduction for 3D range geometry compression

Finley

Bell

2021

Optics and Lasers in Engineering

View full text Add to dashboard Cite

Two-channel depth encoding for 3D range geometry compression

Cited by 9 publications

References 12 publications

SCDeep: Single-Channel Depth Encoding for 3D-Range Geometry Compression Utilizing Deep-Learning Techniques

SCDeep: Single-Channel Depth Encoding for 3D-Range Geometry Compression Utilizing Deep-Learning Techniques

Depth Range Reduction for 3D Range Geometry Compression

Depth range reduction for 3D range geometry compression

Contact Info

Product

Resources

About