About the conferences acceptance rate

In this corner of the ring, the famous IEEE International Conference on Image Processing in its 17th edition, parallel sessions, poster sessions, plenary talks, exhibition hall, so many sponsors that they can't list them, more than a thousand attendees expected, and this year so many submissions that its PC was desperately looking for reviewers from the watermarking community -- even though some of your favorite reviewers didn't get any content security paper to review: that's the sad point in striving (or not being able) to manage such a huge event.

In the opposite corner, the small and dying Information Hiding workshop, single threaded session, no fancy poster, no big head speaking, no exhibition hall, one unique sponsor (Technicolor) and so few submissions that the acceptance rate is likely to be larger than 1/2.

Let us look at previous fights (thanks Google Scholar for the statistics):

• Barni (SPIE 109 cit. , ICIP 87 cit., SPIE 77 cit., SPIE 77 cit.)
• Cachin (IH 425 cit.),
• Cox (IH 347 cit., ICIP 117cit.),
• Delp. (ACM 133 cit., SPIE 123 cit., SPIE 115 cit.)
• Fridrich (SPIE 256 cit., IH 239 cit., IH. 214 cit., ICIP 160 cit., ICIP 100 cit.),
• Kalker (SPIE 199 cit., ICIP 96 cit.)
• Kutter (ICIP 233 cit.)
• Wong (ICIP 310 cit.)

So, which one of two will you attend?

Well, that big foreseen acceptance rate for IH is not very attractive. Some researchers discard such conferences, or are not given credits for their papers published in conference with so big acceptance rate. This is why I say "dying" IH: in our sad times, ROI on advertising might be considered more important than sound research.

HOWEVER, your favorite watermasked cucumbers have been reviewing for both events... and frankly, from what they've read, the IH papers were far, far better.

We do not say that there were all great, but significantly better than ICIP submissions on average (not to say: median). Therefore, judging a conference from its acceptance rate is just as silly as when judgments go on upon only one criterion (notice the alliteration).

Conclusion: if you are solely interested in data hiding and content security, you'd better go to IH. Period.

Miss Cucumber.

Truth hurts

If you want to convince people that digital Forensics can be useful, here is a collection of interesting links.

The most impressive ones are the tampering collection throughout history built by Hany Farid and the museum of Hoaxes.

A smaller and contemporary list can be found on this Spanish web site, and the top ten of doctored photos (a politically correct expression for fakes) presents event obvious fakes.

Tzatziki

Cucumber spirit hits SPIE Optical Engineering

By Cucumber without spiced ham


Ever feared checking your Inbox for a request to review another spam paper? Spam papers in watermarking commonly share the following features:

• The author is not concerned with frivolousnesses like the difference between zero-bit and multi-bit watermarking;
  
• While the technique is not designed to survive geometrical distortions, a 3° rotation generally "demonstrates the high robustness of the proposed approach to geometrical transforms" -- and, of course, pseudo-cropping is always meant for plain regular cropping;
  
• For zero-bit watermarking (which is definitely the same as multi-bit), the threshold is always taken from the Barni et al. paper [1] ensuring a 1e-8 probability of false alarm. No threshold is ever allowed to include another constant than 3.97 which is engraved in the holy Eq. 15 of the said paper. Whether the computation actually applies straightforwardly to the submitted paper is of secondary interest;
• Explicit distortion specification is generally omitted for the sake of simplicity (oh! and Lena looks good anyway when printed on a 2 inch square!);
  
• Security is ensured by the use of a secret key;
  
• Spell-checking is left to the reviewer;
  
• The results always clearly and unconditionally demonstrate the superiority of the proposed method in any area of comparison.
  

It appears however that the times they are a-changin' [2]. A Cucumber of ours recently received the new instructions for reviewers from SPIE Optical Engineering, as part of an invitation to review another spam paper. Hell! These new instructions read:

"Although this paper need not be exceptional, it should add
significantly to the field for you to recommend acceptance or revision.
Lately, a substantial number of papers have been submitted that can be
called "not wrong" papers. These are papers that contain no errors, but
they also lack any new and useful information that would move your field
forward; they may provide no citable results, or document so little
progress that researchers in your field will ignore them. These papers
take up your time and ours; they clutter up the literature; and they do
not advance research in the field. If you find this paper fits this
description, you should recommend that the paper be rejected."

That's pretty good news it finally got written in plain English.

References

[1] M. Barni, F. Bartolini and A. Piva, Improved wavelet-based watermarking through pixel-wise masking, IEEE Trans. Image Proc., vol. 10, issue 5, pp. 783--791, May 2001.
[2] R.A. Zimmerman, The times they are a-changin', Columbia Trans. on Bob Dylan, Special Issue on The Times They Are A-Changin', January 1964.

Distortion-free 3D steganography!

Most of us think as data-hiding as a communication problem with side-information at the encoder. This widely accepted view has led to dramatic improvements of data-hiding techniques over the years. As a direct consequence, any watermark can be seen as a noise that is added to the host content.

These days, a new trend is emerging in 3D data-hiding. Such data is twofold: there is geometry and there is connectivity. Roughly speaking, geometry is a bunch of 3D Cartesian samples, and connectivity is the way the vertices connect to one another. Some people, including several of the most respected in the mesh processing area (see this paper by Bogomjakov, Gotsman and Isenburg) [1], want to hide information in the way the connectivity is described: since geometry is not affected, it leads to distortion-free data-hiding!

To us, it reads like "communications at no power" and other "infinite capacity channel" odd stories. Frightening. Sounds like a definitive breakthrough in information theory.

But wait!

There is a problem. The real question is whether connectivity should be considered useful to describe 3D data. The answer is not easy. There is already a host of works dealing with 3D point clouds data and how to synthesize connectivity from scratch. But there is more: Isenburg has done extremely interesting works on some sort of a "ghost geometry" that is already present only in the connectivity information (see his connectivity shapes).

Turn again to data-hiding as we know it: with regularly sampled host contents. Those ones that we love and that do not need description, except the dimensions of the image or the duration of the song.

Now let's make the following crazy assumption: each and every pixel of an image is to be numbered and transmitted separately. The decoder now has to know the neighborhood of every pixel. Yes. And we do not transmit triangles anymore (like for 3D meshes), but quads. That's images with explicit connectivity.

And finally we can do the same thing: distortion-free data-hiding for images. Sounds weird uh?

Interestingly enough, the paper by Bogomjakov et al. states that one needs to have a reference ordering of the vertices, so the decoder can catch the difference with the transmitted connectivity and compute the hidden message. Similar ideas are shared with so-called permutation watermarking. Therefore, there is actually some sort of distortion in their scheme. BTW, their scheme also has a capacity!

Why people transmitting images do not send quads with YUV/RGB values being the attribute data? That's because everyone assumes this reference ordering of the pixels inside an image. Everyone is happy with YUV/RGB values only.

And since there is this implicit public reference ordering of the vertices and the decoder can catch the difference, an adversary should be able to detect the hidden information quite easily. Then why is it called steganography?

So let's make it clear once for all: these guys do data-hiding on graphs. They don't do distortion-free 3D steganography. For sure.



[1] Although Gotsman already did some sort of Karhunen-Loeve Transform based compression for 3D meshes... Either it's not as real-time as claimed (one needs the decoder to compute the basis vectors -- cost: several O(N^3) diagonalizations of ~500x500 Laplacian matrices), or it is not compression (one needs to transmit the basis vectors). The improvement presented here suffers from a problem apparented to the Gibbs phenomenon.