Abstract

In this paper, we propose an efficient compression method for integral images based on three- dimensional discrete cosine transform (3D-DCT). Even though the existing 3D-DCT based techniques are efficient, they may not be optimized to the characteristics of integral images, such as applying a fixed size block construction and a fixed scanning in placing 2D blocks to construct a 3D block. Therefore, we propose a variable size block construction and a scanning method adaptive to characteristics of integral images, which are realized by adaptive 3D block modes. Experimental results show that the proposed method gives significant improvement in coding efficiency. In particular, at the high bit rates, the proposed method is more improved, since overhead bits for signaling of the 3D block modes take a smaller part of the total bits.

© 2010 Optical Society of America

Full Article  |  PDF Article

References

  • View by:
  • |
  • |
  • |

  1. G. Lippmann, “Epreuves reversible donnant la sensation du relief,” J. Phys. 7, 821–825 (1908).
  2. ITU-T and ISO/IEC JTC 1, “Digital compression and coding of continuous-tone still images,” Recommendation T.81 and ISO/IEC 10 918–1 (1992).
  3. ITU-T and ISO/IEC JTC 1, “Generic coding of moving pictures and associated audio information—Part 2: Video,” Recommendation H.262 and ISO/IEC 13 818–2 (1994).
  4. ITU-T and ISO/IEC JTC 1, “Advanced video coding for generic audiovisual services,” Recommendation H.264 and ISO/IEC 14496–10 (MPEG-4) AVC (2003).
  5. S. Yeom, A. Stern, and B. Javidi, “Compression of 3D color integral images,” Opt. Express 12, 1632–1642 (2004).
    [CrossRef] [PubMed]
  6. N. Sgouros, I. Kontaxakis, and M. Sangriotis, “Effect of different traversal schemes in integral image coding,” Appl. Opt. 47, D28–D37 (2008).
    [CrossRef] [PubMed]
  7. E. Elharar, A. Stern, and O. Hadar, “A hybrid compression method for integral images using discrete wavelet transform and discrete cosine transform,” J. Display Technol. 3, 321–325(2007).
    [CrossRef]
  8. A. Aggoun and M. Tabit, “Data compression of integral images for 3D TV,” in Proceedings of 3DTV Conference (2007), pp. 1–4.
    [CrossRef]
  9. G. J. Sullivan and R. L. Baker, “Efficient quadtree coding of images and video,” IEEE Trans. Image Process. 3, 327–331(1994).
    [CrossRef] [PubMed]
  10. T. Wiegand and B. Girod, “Lagrange multiplier selection in hybrid video coder control,” in Proceedings of International Conference on Image Processing (2001), pp. 542–545.
  11. G. J. Sullivan and T. Wiegand, “Rate-distortion methods for image and video compression,” IEEE Signal Process. Mag. 15, 74 (1998).
    [CrossRef]
  12. http://iphome.hhi.de/suehring/tml/, H.264/AVC reference software (JM).
  13. G. Bjontegaard, “Calculation of average PSNR differences between RD-curves,” ITU-T Q-6/16, Doc.VCEG-M33 (2001).

2008 (1)

2007 (1)

2004 (1)

1998 (1)

G. J. Sullivan and T. Wiegand, “Rate-distortion methods for image and video compression,” IEEE Signal Process. Mag. 15, 74 (1998).
[CrossRef]

1994 (1)

G. J. Sullivan and R. L. Baker, “Efficient quadtree coding of images and video,” IEEE Trans. Image Process. 3, 327–331(1994).
[CrossRef] [PubMed]

1908 (1)

G. Lippmann, “Epreuves reversible donnant la sensation du relief,” J. Phys. 7, 821–825 (1908).

Aggoun, A.

A. Aggoun and M. Tabit, “Data compression of integral images for 3D TV,” in Proceedings of 3DTV Conference (2007), pp. 1–4.
[CrossRef]

Baker, R. L.

G. J. Sullivan and R. L. Baker, “Efficient quadtree coding of images and video,” IEEE Trans. Image Process. 3, 327–331(1994).
[CrossRef] [PubMed]

Bjontegaard, G.

G. Bjontegaard, “Calculation of average PSNR differences between RD-curves,” ITU-T Q-6/16, Doc.VCEG-M33 (2001).

Elharar, E.

Girod, B.

T. Wiegand and B. Girod, “Lagrange multiplier selection in hybrid video coder control,” in Proceedings of International Conference on Image Processing (2001), pp. 542–545.

Hadar, O.

Javidi, B.

Kontaxakis, I.

Lippmann, G.

G. Lippmann, “Epreuves reversible donnant la sensation du relief,” J. Phys. 7, 821–825 (1908).

Sangriotis, M.

Sgouros, N.

Stern, A.

Sullivan, G. J.

G. J. Sullivan and T. Wiegand, “Rate-distortion methods for image and video compression,” IEEE Signal Process. Mag. 15, 74 (1998).
[CrossRef]

G. J. Sullivan and R. L. Baker, “Efficient quadtree coding of images and video,” IEEE Trans. Image Process. 3, 327–331(1994).
[CrossRef] [PubMed]

Tabit, M.

A. Aggoun and M. Tabit, “Data compression of integral images for 3D TV,” in Proceedings of 3DTV Conference (2007), pp. 1–4.
[CrossRef]

Wiegand, T.

G. J. Sullivan and T. Wiegand, “Rate-distortion methods for image and video compression,” IEEE Signal Process. Mag. 15, 74 (1998).
[CrossRef]

T. Wiegand and B. Girod, “Lagrange multiplier selection in hybrid video coder control,” in Proceedings of International Conference on Image Processing (2001), pp. 542–545.

Yeom, S.

Appl. Opt. (1)

IEEE Signal Process. Mag. (1)

G. J. Sullivan and T. Wiegand, “Rate-distortion methods for image and video compression,” IEEE Signal Process. Mag. 15, 74 (1998).
[CrossRef]

IEEE Trans. Image Process. (1)

G. J. Sullivan and R. L. Baker, “Efficient quadtree coding of images and video,” IEEE Trans. Image Process. 3, 327–331(1994).
[CrossRef] [PubMed]

J. Display Technol. (1)

J. Phys. (1)

G. Lippmann, “Epreuves reversible donnant la sensation du relief,” J. Phys. 7, 821–825 (1908).

Opt. Express (1)

Other (7)

ITU-T and ISO/IEC JTC 1, “Digital compression and coding of continuous-tone still images,” Recommendation T.81 and ISO/IEC 10 918–1 (1992).

ITU-T and ISO/IEC JTC 1, “Generic coding of moving pictures and associated audio information—Part 2: Video,” Recommendation H.262 and ISO/IEC 13 818–2 (1994).

ITU-T and ISO/IEC JTC 1, “Advanced video coding for generic audiovisual services,” Recommendation H.264 and ISO/IEC 14496–10 (MPEG-4) AVC (2003).

A. Aggoun and M. Tabit, “Data compression of integral images for 3D TV,” in Proceedings of 3DTV Conference (2007), pp. 1–4.
[CrossRef]

T. Wiegand and B. Girod, “Lagrange multiplier selection in hybrid video coder control,” in Proceedings of International Conference on Image Processing (2001), pp. 542–545.

http://iphome.hhi.de/suehring/tml/, H.264/AVC reference software (JM).

G. Bjontegaard, “Calculation of average PSNR differences between RD-curves,” ITU-T Q-6/16, Doc.VCEG-M33 (2001).

Cited By

OSA participates in CrossRef's Cited-By Linking service. Citing articles from OSA journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (8)

Fig. 1
Fig. 1

Two possible assemblies of elemental images for 3D-DCT. (a) Horizontal grouping. (b) Horizontal-vertical grouping. (c) A 3D block constructed by grouping.

Fig. 2
Fig. 2

Examples of integral images. (a) An image with a banana object. (b) An image with an apple and a banana objects.

Fig. 3
Fig. 3

Basic units for compression. A bold solid box is a basic unit.

Fig. 4
Fig. 4

Scanning methods in a basic unit. (a) Horizontal scan mode. (b) Vertical scan mode.

Fig. 5
Fig. 5

Structure of a basic unit that contains multiple blocks in a single elemental image ( N = 24 ).

Fig. 6
Fig. 6

Four block modes in the x y image plane. (a) 8 × 8 block mode. (b) 8 × 4 block mode. (c) 4 × 8 block mode. (d) 4 × 4 block mode.

Fig. 7
Fig. 7

Three images for evaluation. (a) Test image 1. (b) Test image 2. (c) Test image 3. (d) Objects used for rendering three test images.

Fig. 8
Fig. 8

PSNR versus the number of bits. (a) Test image 1. (b) Test image 2. (c) Test image 3.

Tables (2)

Tables Icon

Table 1 Coding Results According to a Few Fixed Block Sizes

Tables Icon

Table 2 Performance Comparison of the Proposed Method and the Conventional Methods

Equations (5)

Equations on this page are rendered with MathJax. Learn more.

S ( u , v , w ) = c ( u , v , w ) z = 0 N z 1 y = 0 N y 1 x = 0 N x 1 s ( x , y , z ) cos ( 2 x + 1 ) u π 2 N x cos ( 2 y + 1 ) v π 2 N y cos ( 2 z + 1 ) w π 2 N z ,
s ( x , y , z ) = w = 0 N z 1 v = 0 N y 1 u = 0 N x 1 c ( u , v , w ) S ( u , v , w ) cos ( 2 x + 1 ) u π 2 N x cos ( 2 y + 1 ) v π 2 N y cos ( 2 z + 1 ) w π 2 N z ,
c ( u , v , w ) = c ( u ) c ( v ) c ( w ) s N x N y N z , c ( k ) = { 1 / 2 , k = 0 1 , k 0 .
J ( MODE | Q ) = D ( MODE | Q ) + λ ( Q ) R ( MODE | Q ) ,
MODE opt = arg min MODE J ( MODE | Q ) .

Metrics