●2018 2nd International Symposium on Intelligent Unmanned Systems and Artificial Intelligence(SIUSAI 2018)-- Ei Compendex & Scopus—Call for papers
｜August 27-29, 2018. ｜Las Vegas, USA｜Website: http://www.siusai.org/
●SIUSAI 2018 welcomes researchers, engineers, scientists and industry professionals to an open forum where advances in the field of Intelligent Unmanned Systems and Artificial Intelligence can be shared and examined. The conference is an ideal platform for keeping up with advances and changes to a consistently morphing field.
●Publication and Indexing
All accepted papers will be published in the digital conference proceedings which will be sent to be Indexed by all major citation databases such as Ei Compendex, SCOPUS, Google Scholar, Cambridge Scientific Abstracts (CSA), Inspec, SCImago Journal & Country Rank (SJR), EBSCO, CrossRef, Thomson Reuters (WoS), etc.
A selection of papers will be recommended to be published in international journals.
●Program Preview/ Program at a glance
August 27, 2018: Registration + Icebreaker Reception
August 28, 2018: Opening Ceremony+ KN Speech+ Technical Sessions
August 29, 2018: Technical Sessions+ Half day tour/Lab tours
1. PDF version submit via CMT: https://cmt3.research.microsoft.com/User/Login?ReturnUrl=%2Fsiusai2018
2. Submit Via email directly to: firstname.lastname@example.org
Fast compressed domain image retrieval
Loughborough University, UK
Speakers - ICVISP 2018
Content-based image retrieval (CBIR) has been an active research area for many years. However, almost all images are stored in compressed form, the vast majority of CBIR algorithms operate in the (uncompressed) pixel domain. This not only leads to a computational overhead for feature calculation, it can also be shown that image compression affects retrieval accuracy, especially at extreme compression rates.
In my talk, I will discuss efficient and effective CBIR techniques that operate directly in the compressed domain. In particular, I will focus on JPEG compressed images since most images are stored in this format. Our compressed domain retrieval techniques eliminate the need of full decompression for feature extraction while matching common pixel domain methods in terms of retrieval performance. Last not least I will show how retrieval can be achieved without any decompression at all, by exploiting adapted information stored in the header of JPEG files.