ICIP 2010 - Feb. 8

I thought the deadline for ICIP 2010 is today, but a quick look at the website, it's been extended for two weeks, which suggests that the economy in the research community hasn't completely recovered yet, and North American institutions are still struggling to find money to get students/prof to go oversea to Asia for a conference.

And this is also an opportunity for me to take advantage of the extension to work hard and get a chance to go back to my motherland.

GPU - future of computing

My prof thinks GPU will play a huge role in the future of computing, especially in image/video coding. Real-time encoding of video has always been an issue due to the complexity of the encoder, especially with H.264 being a much more sophisticated encoder than its predecessor. But with the advent of GPU processing, these encoder complexity will become lesser of an issue, assuming that one can achieve parallel programming at the encoder side.

I think the future of video coding must make use of parallel coding paradigm. Maybe the design of such codecs may not give you optimal performance, but it should be aware of what parallel computing can offer, and thus strike a balance between performance and complexity. After all, this is always the main tradeoff in computing, isn't it?

acm programming from SLPC

The Stanford Local Programming Contest page has a lot of goodies (resources) provided to its students, or everyone really. I went nuts last night downloading most of the problems.

http://cs.stanford.edu/group/acm/SLPC/

Of course the real key is to read the problems and think about them, and knowing that they provided solutions really helped because you can improve by comparing your solution and *the* solution.

I am thinking these problems can occupy my time during the times when I do my daily commute from point A to point B, as well as being physically at places where mentally I can be doing something else.

Introduction to Algorithms

I feel a bit ashamed that I'm starting to read this book now, when I should have been reading it 10 years ago. So when I found out that the 3rd edition came out not too long ago, I purchased the book out of impulse response. I guess when a book is regarded as a bible in the field, then it's ok to own it.

http://www.amazon.ca/Introduction-Algorithms-Third-Thomas-Cormen/dp/0262033844/ref=pd_sim_b_1

video lectures

Video lectures are found almost everywhere now. I really like the idea of sharing knowledge this way. It really lets students who are not from that university to get "free" education, although I think the money is really on the office hours, feedback, and learning environment that the teaching staff provides.

Anyway, I want to take advantage of those resources and learn from two good Stanford profs in this quarter:

http://viewxtreme.stanford.edu/

Geoffrey A. Moore's books

PIcked up a few books from Moore from the library, and looking forward to reading it and picking up some general knowledge.

Crossing the Chasm
The Gorilla Game
Inside the Tornado
Living on the fault line

google i/o 2010

Had a sudden urge to register for Google I/O 2010 on May 9-10, 2010.

http://code.google.com/events/io/2010/

However, I will probably take a pass and watch the sessions on youtube later on.

toc-alert - Jan 11, 2010

Circuits and Systems for Video Technology, IEEE Transactions on
Volume: 20, Issue: 1

A couple of papers from this journal caught my eye:

Rate-Distortion Cost Estimation for H.264/AVC, by Liao, K.-Y.; Yang, J.-F.; Sun, M.-T.

I have met Prof M.T. Sun at PV 2009 last year, but didn't know who he was then. I didn't even know he compiled the book "Compressed Video over Networks", otherwise I would have talked to him. This paper is interesting because it presented a new way to estimate the RD function for H.264 encoding. It is an upgrade of an earlier method that is based on the CAVLC entropy coding scheme. I haven't read it seriously enough to say more...



Image Processing, IET
Volume: 4 Issue: 1 Date: February 2010

Statistical motion learning for improved transform domain Wyner-Ziv video coding, by Martins, R.; Brites, C.; Ascenso, J.; Pereira, F.

I've followed this group quite closely before, reading most of their DVC papers. Now it seems like they are using machine learning concepts for their turbo-code based DVC. I'll definitely take a look at this paper when I have time.