Scribd
Upload
72 views1 page

Efficient Color Compression via Diffusion

The document discusses turning colorization methods into efficient color compression. It proposes a new luma-guided diffusion framework that steers color channel inpainting according to brightness channel structure. This outperforms previous colorization methods. A new luma preference codec invests most bits into accurate brightness representation. This allows high quality color reconstruction with the colorization technique, while exploiting that humans are more sensitive to structure than color. Experiments show the new codec outperforms state-of-the-art diffusion compression and competes with transform codecs.

Uploaded by

Anonymous 1aqlkZ
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
72 views1 page

Efficient Color Compression via Diffusion

The document discusses turning colorization methods into efficient color compression. It proposes a new luma-guided diffusion framework that steers color channel inpainting according to brightness channel structure. This outperforms previous colorization methods. A new luma preference codec invests most bits into accurate brightness representation. This allows high quality color reconstruction with the colorization technique, while exploiting that humans are more sensitive to structure than color. Experiments show the new codec outperforms state-of-the-art diffusion compression and competes with transform codecs.

Uploaded by

Anonymous 1aqlkZ
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

TURNING DIFFUSION-BASED IMAGE COLORIZATION INTO

EFFICIENT COLOR COMPRESSION

Abstract:
The work of Levin et al. (2004) popularized stroke-based methods that add color
to gray value images according to a small amount of user-specified color samples. Even
though such reconstructions from sparse data suggest a possible use in compression, only
few attempts were made so far in this direction. Diffusion-based compression methods
pursue a similar idea: they store only few image pixels and inpaint the missing regions.
Despite this close relation and a lack of diffusion-based color codecs, colorization ideas
were so far only integrated into transform-based approaches such as JPEG. We address
this missing link with two contributions. First, we show the relation between the discrete
colorization of Levin et al. and continuous diffusion-based inpainting in the YCbCr color
space. It decomposes the image into a luma (brightness) channel and two chroma (color)
channels. Our luma-guided diffusion framework steers the diffusion inpainting in the
chroma channels according to the structure in the luma channel. We show that making
the luma-guided colorization anisotropic outperforms the method of Levin et al.
significantly. Second, we propose a new luma preference codec that invests a large
fraction of the bit budget into an accurate representation of the luma channel. This allows
a high-quality reconstruction of color data with our colorization technique.
Simultaneously, we exploit the fact that the human visual system is more sensitive to
structural than to color information. Our experiments demonstrate that our new codec
outperforms the state of the art in diffusion-based image compression and is competitive
to transform-based codecs.

Proposed System :
We  evaluate  four  different diffusion models from our general framework,
namely space-variant  isotropic  and  space-variant  anisotropic  diffusion,  as well  as 
two  newly  proposed  higher-order  counterparts  to  the aforementioned  models.  We 
show  that  the  discrete  method of  Levin  et  al.  is  closely  related  to  our  continuous 
higher-order isotropic diffusion. Our experiments on well-known test images 
demonstrate  that  our  anisotropic  models  outperform isotropic methods significantly.

You might also like