From Mr Tsukemono,
Mr. Tsukemono was sad to see only 4 papers related to Steganography and Steganalysis on the whole conference. The general trend of the decreasing number of papers continues, as WIFS enjoyed only one such paper. But the good news is that the quality of papers was good. And quality is preferred over quantity.
Mr. Tsukemono has enjoyed following two papers:
- Application of Viterbi algorithm to asymptotically reach rate distortion bound due to T. Filler, J. Judas and J. Fridrich.
- YASS algorithm is finally broken by blind steganalysis, though 1200 features makes Mr. Tsukemono rather bitter.
Remark: The session was dominated by Binghamton University, from which 3 out of 4 papers came. What are the other groups doing? Are they saving papers for IHW, ACM, WIFS, IWDW, ICIP, ICASP or what else? We should hope. By considering number of papers and sessions, forget steganography and steganalysis, forensics is the hype now. Forensics was subject of 5 sessions. We could see great papers coming from Dartmouth, Dresden, Binghamton, Purdue, and others. It was not coincidence that 3 papers dealt with the same topic, i.e. how to efficiently search and match fingerprint in large database. Each paper approached the problem from completely different direction.
Another important point: we could finally see an attempt to plant a new camera fingerprint to the image (and method to detect this forgery). The watermarking is getting popular as well. The number of accepted papers was so large that the papers had to be presented in parallel sessions. Good for watermarking, bad for conference attendees. Although tutorial talks this year lacked stunning pictures from deep sea or from deep inside of the human body, they made Mr. Tsukemono a goose flesh. He realized that his new shiny iPhone is spying on him...
