Proceedings of the 2017 ACM on International Conference on Multimedia Retrieval
Alexander G. Hauptmann,Chong-Wah Ngo,Xiangyang Xue,Yu-Gang Jiang,Cees G. M. Snoek,Nuno Vasconcelos
2015-01-01
Abstract:It is our great pleasure to welcome you to the 2017 ACM International Conference on Retrieval (ICMR 2017). This year's conference continues a 17-year tradition of being the premier forum for presentation of research results and experience reports on core topics in multimedia retrieval, as well as the broader set of topics that must be addressed to ensure that multimedia retrieval technologies are of practical use in real-world use cases. Special emphasis was placed on topics related to large-scale indexing, user interaction, exploiting diverse and multimodal data, and domain-specific challenges. This annual international conference, combines the long-standing experience of the former ACM CIVR (International Conference on Image and Retrieval) and ACM MIR (International Conference on Retrieval) series, and was set up in 2011 to highlight the state of the art in multimedia (e.g., text, image, video, audio, sensor data, and 3D) retrieval.
This year's edition has attracted submissions from all over the world: Asia, Canada, Australia, Africa, Europe, and the United States. In total, it received 178 valid submissions, which were reviewed by the program committee. For the classic research paper tracks, out of 95 full paper submissions, 33 were accepted -- 20 for full oral presentation and 13 for spotlight oral presentation (acceptance rate of 34%; full oral paper acceptance rate of 21%) -- and, out of 35 short paper submissions, 13 were accepted (acceptance rate of 37%).
Next to full and short paper tracks, the conference includes a diversity of other tracks: open-source software (acceptance rate of 80%), technical demonstrations (acceptance rate of 85%), brave new ideas papers (acceptance rate of 75%), doctoral symposium papers (acceptance rate of 80%), and special session papers (acceptance rate of 66%). In addition to the tracks of the main conference, two workshops are offered: Wearable MultiMedia and Multimedia Forensics and Security.
The conference program is complemented by a full day of tutorials on Video Indexing, Search, Detection, and Description, keynote talks given by world-class researchers (Searching for a thing, Arnold W.M. Smeulders & Ran Tao, and Making cultural visits with a smart mate, Alberto Del Bimbo), industry keynote talks given by leading companies (With 5G Approaching, How will Audio/Video Technology that Serves 800 Million QQ Users Bring Forth New Ideas, Xiaozheng Huang, Tencent, China, Information Retrieval from Multi-Sensor Data for Enriching Location Services at HERE Technologies, Matei Stroila, HERE, USA, and Intelligently Connecting People with Information, Changhu Wang, Toutiao AI Lab, China) and a panel session on Indicators of Innovation and Signs of Success: Recognizing which multimedia information retrieval research is most valuable to industry.