About

The 12th Asian Conference on Computer Vision (ACCV 2014) will be held in Singapore on Nov 1-5, 2014. ACCV is a leading biennial international conference focused on computer vision, pattern recognition, and related areas. This highly successful series provides a premier forum for researchers, developers, and practitioners to present and discuss new problems, solutions, and technologies in computer vision and related areas. ACCV 2014 solicits high-quality original research for publication in its main conference and co-located workshops.

Call for ACCV'16 proposals - Due Oct 20, 2014 (link)

Important Dates

Paper Submissions June 20, 2014 - 23:59 PDT (GMT-7)
Supp Materials Submissions June 27, 2014 - 23:59 PDT (GMT-7)
Rebuttal Aug 16-21, 2014 - 23:59 PDT (GMT-7)
Author Notification Sep 5, 2014 - 23:59 PDT (GMT-7)
Camera Ready Oct 1, 2014 - 23:59 PDT (GMT-7)
Conference Nov 1-5, 2014

 

The complete schedule for ACCV 2014 is now available, see program section of the webpage (link). If you haven't already, please book your hotel and finalize your travel arrangements as soon as possible. Details on transportation in Singapore and the conference can be found here (link).

Award papers will be invited to submit a journal version to a special issue of Computer Vision and Image Understanding (CVIU). Outstanding applications papers will be invited to submit a journal version to IJSJ Computer Vision Applications.


For Presenters (Posters and Orals)

 

Poster Presentations

The ACCV 2014 poster boards are 1 meter by 1 meter. We strongly recommend that you use posters size is A1 - [559mm x 841mm] (see PPT file link). We recommend that you prepare your poster in portrait orientation (height=841mm, width=559mm). Poster sessions are held each day of the main conference from 1.15pm -3.15pm. Presenters are requested to setup their posters in the morning before the keynote and first oral session. We have several tables available for poster presenters that are available upon request. Please contact a student volunteer during setup time to get a table.

Oral Presentations

Speakers at the main conference will be given 15 minutes each. Speakers should prepare a 12-13 minute talk and leave 2-3 mintues for Q&A. Speakers are requested to come to the podium 20 minutes before the start of their oral session to test their laptops and slides. Speakers at workshops should contact the workshop organizers regarding their presentation duration. We do have any required slide templates, but the ACCV 2014 logo can be found here if you would like to include it in your presentation (link). The conference site informed us that that projector is a standard XGA (1024x768) with 4:3 aspect ratio.


For Authors

 

Instructions for preparing camera ready paper
(Main conference and workshops)


Proceedings
Accepted papers will be published after the conference in four volumes of Springer's Lecture Notes on Computer Science (LNCS). During the conference, each conference participant will receive an e-copy of all the accepted papers, including supplemental videos as submitted by authors with their final paper (as approved by the program chairs).

Submission Site
Main conference authors can upload their final papers using the main CMT submission site ( link )
Please ensure that you carefully follow the procedure below before uploading your paper.

Submissions
Camera ready papers can be up to 14 pages (excluding references). A final paper submission has to provide the following:
  1. The full Latex sources include figures in PDF, PNG or JPG format. This should be all in one directory.
  2. A signed copyright form (the form is in the author camera ready kit below).
  3. A PDF of the paper (maximum 10MB in LNCS style and within the require page length).
  4. A text file named "contact.txt" containing the author contact information (see below for the format).
Authors may also submit additionally one supplementary material. Submissions are via the ACCV 2014 submission website (https://cmt.research.microsoft.com/ACCV2014/).
All files for one paper need to be zipped into one archive.
Final papers will be compared to submitted papers. A violation of one of the above items could lead to rejection of your paper.

Registration, Publication Fee, and Invitation Letters for Visa
Registration At least one author should register by October 1st (deadline of camera ready) to have your paper included in the proceedings. A full conference registration can cover multiple papers only if the registered person is an author on the paper. Regular and student registrations can cover papers accepted to both the main conference and workshops. However, workshop registration is can only be used for papers accepted at a workshop. If you need a visa letter you can request this as part of the registration procedure.

Publication Fee Previous ACCVs had extra page charges, however, for ACCV 2014 all accepted papers to the main conference will be charged a flat fee of SGD$150. This fee is required for all accepted main conference papers.
Conference registration and publication fee payment sites are available below under Registration.

Camera Ready Author Kit
The author kit provides a LaTeX2e template for the final paper submissions, the LNCS copyright form, and sources of an example paper to demonstrate the format. Please refer to this example for detailed formatting instructions.

Camera ready author kit as a zip file ( link )

Instructions and Policies
Final camera ready (due Oct 1st) must be prepared in LaTeX2e, thus suitable for compiling the proceedings in one uniform text editing environment. This is a requirement by LNCS. Any paper not prepared in Latex2e using the supplied camera ready kit cannot be accepted.

Author names : Please write out author names in full in the paper (full first and family names).

Copyright form : One copyright form per paper has to be provided. The author kit contains the template of the form. Please fill in the form and sign it (the signature of corresponding author is sufficient). Either a digitally signed or scanned version of the copyright form (PDF format) needs to be submitted together with the final paper. Please use the appropriate copyright form for main conference or workshop papers (there are two forms in the camera ready kit).

All source files : Latex2e files for the text, PS/EPS or PDF/PNG/JPEG files for all figures need to be submitted. We should be able to compile your camera-ready pdf from directly from your camera-ready submission. Please prepare your tex file such that your paper can be built in a single directory (i.e. do not place figures in subdirectories).

Contact file : A text file named "contact.txt" containing the following author information needs to be submitted.
Name and email address of a corresponding author (author who will be contacted for proof-reading).

For each author, in the order in which authors appear on the paper: Full family name(s), full first name(s), email address(es), affiliation(s) (name of the institution as it should appear in the conference program, not the full address). Please separate these 4 items by ",".

Supplemental material (Optional): Authors may optionally upload one supplemental material, e.g., video, also zipped into the single archive.

Overall file size: The above needs to be submitted in a single ZIP file to ACCV 2014 submission website. This archive may be 30MB at most. Here is the check list of files that need to be contained in the ZIP file.



Contact
Questions regarding the final paper upload or copyright form should be addressed to the publication chair: Jianxin Wu (wujx2001 {AT} gmail.com) and cc the program chairs at (programchairs {AT} accv2014.org).

No show policy
Every accepted paper to ACCV 2014 (both main conference and workshop) needs to have at least one author attend the conference and represent the paper in either the poster or oral session. The ACCV 2014 organizing committee reserves the right to remove a paper from the final proceedings that was not represented at the conference. No show authors will not have their registrations refunded.


Organizers

 

General Chairs

Michael S. Brown, National University of Singapore, Singapore
Tat-Jen Cham, Nanyang Technological University, Singapore
Yasuyuki Matsushita, Microsoft Research Asia, China

Program Chairs

Daniel Cremers, Technical University of Munich, Germany
Ian Reid, University of Adelaide, Australia
Hideo Saito, Keio University, Japan
Ming-Hsuan Yang, University of California Merced, USA

Organizing Chairs

Teck Khim Ng, National University of Singapore, Singapore
Junsong Yuan, Nanyang Technological University, Singapore

Workshop Chairs

CV Jawahar, IIIT-Hyderabad, India
Shiguang Shan, Institute of Computing Technology, Chinese Academy of Sciences, China

Demo Chairs

Bohyung Han, POSTECH, Korea
Koichi Kise, Osaka University, Japan

Tutorial Chairs

Chu-Song Chen , Academia Sinica, Tawain
Brendan McCane, Otago University, New Zealand

Publication Chairs

Terence Sim, National University of Singapore, Singapore
Jianxin Wu, Nanjing University, China

Steering Committee

In-So Kweon , KAIST, Korea
Yasushi Yagi , Osaka University, Japan
Hongbin Zha, Peking University, China

Honorary Chair

Katsushi Ikeuchi, University of Tokyo, Japan

Area Chairs

Lourdes Agapito
Thomas Brox
Tat-Jun Chin
Yung-Yu Chuang
Larry Davis
Yasutaka Furukawa
Bastian Goldluecke
Bohyung Han
Hiroshi Ishikawa
C.V. Jawahar
Jana Kosecka
David Kriegman
Shang-Hong Lai
Ivan Laptev
Kyoung Mu Lee
Vincent Lepetit
Jongwoo Lim
Simon Lucey
Ajmal Mian
Hajime Nagahara
Ko Nishino
Shmuel Peleg
Imari Sato
Shin'ichi Satoh
Stefano Soatto
Jamie Shotton
Ping Tan
Lorenzo Torresani
Manik Varma
Xiaogang Wang
Shuicheng Yan
Qing-Xiong Yang
Jingyi Yu
Junsong Yuan
Hongbin Zha
Lei Zhang
QMUL/UCL , UK
University of Freiburg , Germany
Adelaide , Australia
National Taiwan University , Taiwan
Maryland , USA
Washington University in St. Louis , USA
University of Konstanz , Germany
POSTECH , Korea
Wasda Univ. , Japan
IIIT Hyderabad , India
George Mason University , USA
UCSD , USA
National Tsing-Hua U , Taiwan
INRIA Rocquencourt , France
Seoul National U , Korea
EPFL , Switzerland
Hanyang U , Korea
CSIRO/UQ , Australia
UWA , Australia
Kyushu University , Japan
Drexel , USA
Hebrew University , Israel
NII , Japan
NII , Japan
UCLA , USA
MSR , UK
SFU , Canada
Dartmouth College , USA
Microsoft Research , India
CUHK , Hong Kong
NUS , Singapore
City University of Hong Kong , Hong Kong
University of Delaware, USA
NTU , Singapore
Peking University , China
HK Polytechnic U , Hong Kong


Outstanding Reviewers


We would like to thank the following reviewers for their outstanding service in reviewing papers submitted to ACCV 2014. Outstanding reviewers were nominated from over 600 reviewers by Area Chairs based on the quality of their reviews.

Emre Akbas Jonathan Balzer Steve Branson Sanderson Conrad Marco Cristani
Alessio Del Bue Anthony Dick Bruce Draper Katerina Fragkiadaki Tatsuya Harada
Mehrtash Harandi Nazli Ikizler-Cinbis Catalin Ionsecu Suha Kwak Junseok Kwon
Fuxin Li Chen-Change Loy Scott McCloskey Xing Mei Yasushi Mikihara
Guy Rosman Mathieu Salzmann Pramod Sankar Walter Scheirer Bernt Schiele
Chunhua Shen Sudipta Sinha Deqing Sun Yuichi Taguchi Toru Tamaki
Dong Wang Yu-Chiang Frank Wang Paul Wohlhart John Wright Bangpeng Yao


Program


Program Schedule


The full ACCV 2014 conference program schedule is now available (click image for the PDF). This program guide includes workshop and tutorial schedules. Please contact workshop organziers directly for questions concerns the workshop schedules. Printed booklets will be given to conference participants at the registration counter.

Click here for details on transportation to the venues (link).

Click here for details on food options at the workshops/tutorials and additional information about Singapore (link).


Keynote Speakers


We are honored to have the following three keynote speakers at ACCV 2014.


Stéphane Mallat
École Normale Supérieure
Title: Deep Convolution Networks as Geometric Image Representations

Until recently, computer vision algorithms mostly resulted from geometric considerations over shapes, textures, motion and 3D perspective projections. Deep learning algorithms seem to ignore these geometric considerations, while providing state of the art classification results on complex image data bases. It is thus time to wonder what type of image information is extracted by these deep neural networks and why do they work so well.

We show that deep convolution networks provide remarkable architectures to represent geometric image information, including shapes, textures and complex structured objects. They compute descriptors having appropriate invariance and stability to geometric transformations. The properties of these deep representations have strong similarities with Gestalt perception. However, the network filters do not need to be learned and can adapted to prior information on geometry, with multiscale wavelets. This brings us back from learning everything to more traditional computer vision approaches. Classification results will be shown on multiple image data bases, with linear SVM applied to network outputs.



Minoru Etoh
NTT Docomo

Title: How Changing Mobile and Media Technologies is Changing The Way We Create Innovations

According to Schumpeter's definition of "Innovation," all the innovation instances are combinations of technologies that already exist. In that context, this talk covers the combination of progress of mobile network technologies and media understanding technologies. When the first mobile phones of 2nd Generation came out, people thought it was only for speech communication and texting. Then the phones got cameras, GPS. accelerometers, near filed communication devices together with fat communication pipes, people in all walks of life started using mobile phones in various opportunities. It is noteworthy that the progress of media understanding applications remarkably creates big data in a way of virtuous cycle by service and technology developments. This talk also highlights big-data driven service developments and API strategies for mobile innovations as well as technologies.



Dieter Fox
University of Washington

Title: RGB-D Perception in Robotics
RGB-D cameras provide per pixel color and depth information at high frame rate and resolution. Gaming and entertainment applications such as the Microsoft Kinect system resulted in the mass production of RGB-D cameras at extremely low cost, also making them available for a wide range of robotics applications. In this talk, I will provide an overview of depth camera research done in the Robotics and State Estimation Lab over the last five years. This work includes 3D mapping, autonomous object modeling, unsupervised feature learning for object recognition, and articulated object tracking.

Workshops

Workshops will be held on Nov 1-2, 2014 at the Stephen Riady Centre on the National University of Singapore campus.

The following workshops will be co-located with ACCV 2014.
Please see the link of each workshop for more information.


Workshop papers will be published together with ACCV 2014 as a volume (titled ACCV 2014 Workshops) in the Springer Lecture Notes on Computer Science (LNCS) series. Please use the same camera ready kit for preparing workshop papers as main conference papers (instructions are listed above), however, please sign the workshop copyright form. Also note that workshop papers do not need to pay a publication fee.

1. Human Gait and Action Analysis in the Wild: Challenges and Applications ( webpage )
    Nov 1, (half day, Morning)
2. 2nd International Workshop on Big Data in 3D Computer Vision ( webpage )
    Nov 1, (all day)
3. Deep Learning on Visual Data ( webpage )
    Nov 2, (half day, afternoon)
4. Workshop on Scene Understanding for Autonomous Systems (SUAS) ( webpage )
    Nov 2, (all day)
5. (Canceled) Large-scale Scene Reconstruction and Visual Mapping using Crowdsourcing Images ( webpage )
    Nov 1, (half day, afternoon) Contact workshop organizers for details.
6. RoLoD: Robust Local Descriptors for Computer Vision ( webpage )
    Nov 1, (all day)
7. Emerging Topics In Image Restoration and Enhancement ( webpage )
    Nov 1, (all day)
8. 1st International Workshop on Robust Reading ( webpage )
    Nov 2, (all day)
9. 2nd Workshop on User-Centred Computer Vision (UCCV 2014) ( webpage )
    Nov 1, (all day)
10. International Workshop on Video Segmentation in Computer Vision ( webpage )
    Nov 2, (all day)
11. My Car Has Eyes - Intelligent Vehicles with Vision Technology ( webpage )
    Nov 2, (all day)
12. Feature and Similarity Learning for Computer Vision ( webpage )
    Nov 1, (all day)
13. 3rd ACCV Workshop on e-Heritage ( webpage )
    Nov 2, (all day)
14. 3rd International Workshop on Intelligent Mobile and Egocentric Vision (IMEV2014) ( webpage )
    Nov 2, (all day)
15. Computer Vision for Affective Computing (CV4AC) ( webpage )
    Nov 1, (all day)
16. Workshop on Human Identification for Surveillance (HIS) ( webpage )
    Nov 2, (all day)


Tutorials

Tutorials will be held on Nov 1-2, 2014 at the Stephen Riady Centre on the National University of Singapore campus.

We are happy to announce the following tutorials have been accepted to appear at ACCV 2014. Click on links for more details.

1. Tutorial on Color Transfer, Ferradans and Bertalm ( link )
    Nov 1, (half day)
2. Advanced Sparse Representation Models for Image and Video Analysis, Gao, Jia, Zhang, and Dong ( link )
    Nov 2, (half day)
3. Tutorial on Riemannian Geometry in Computer Vision, Porikli, Harandi, Sanderson ( link )
    Nov 1, (half day)
4. Essence of Geometric Algebra, Kanatani ( link )
    Nov 2, (half day)
5. Metric Learning for Visual Recognition, Lu, Wang, Deng ( link )
    Nov 2, (half day)
6. Mining Image and Video Data, Yuan and Wu ( link )
    Nov 1, (half day)

Demos

Demos will be held on Nov 3-5, 2014 during the main conference poster sessions.

We are happy to announce the following demos have been accepted to appear at ACCV 2014.

1. Robust Real Time 6-DOF Tracking via Object Coordinate Regression
Alexander Krull, Frank Michel, Eric Brachmann, Stefan Gumhold, Stephan Ihrke, and Carsten Rother (Technical University Dresden)
2. "i-Bolt": Image-based Individual Identification of Metal Parts without Tag
Rui Ishiyama and Toru Takahashi (NEC Corporation)
3. Efficient Image Detail Mining
Andrej Mikulik, Filip Radenovic, Ondrej Chum, and Jiri Matas (Czech Technical University in Prague)
4. Novel Continuous-Multi-Class Cascade for Real-Time Emotional Recognition
Jinhui Chen, Tetsuya Takiguchi and Yasuo Ariki (Kobe University)
5. Scyllarus: A Matlab toolbox aimed at supporting the research on imaging spectroscopy for scene analysis
Nariman Habili, Ran Wei and Antonio Robles-Kelly (NICTA, Australian National University)
6. Automatic Real Crime Scene Shoeprint Retrieval System
Xinnian Wang, Huihui Sun, Qing Yu and Chi Zhang (Dalian Maritime University)
7. Stereo Fusion using a Refractive Medium on a Binocular Base
Seung-Hwan Baek and Min H. Kim (KAIST)
8. VideoSum: A Video Storing, Processing and Summarization Platform
Vasileios Chasanis, Costas Voglis Antonios Ioannidis, Aristidis Likas (University of Ioannia), Aris Lanaridis, Eleni Vathi, Geogios Siolas, and Andreas Stafylopatis (University of Athens)
9. Audio and Video Surveillance System for Public Safety
Takeshi Arikuma, Masahiro Tani and Tsunehisa Kawamata (NEC Labatorites Singapore)

Registration

Registration Info


Category Early Registration
(on/before Oct 5)
Pre-registration
(on/before Oct 20)
Late/On-site Registration**
(after Oct 25)
Regular Main Conference   SGD 850 (~USD 675) SGD 950 (~USD 750) SGD 1100 (~USD 875)
Student Main Conference   SGD 550 (~USD 435) SGD 650 (~USD 520) SGD 800 (~USD 640)
Workshop Only   SGD 250 (~USD 199) SGD 300 (~USD 238) SGD 350 (~USD 278)


Regular and student main conference registrations includes access to the workshops/tutorials, the main conference sessions (Nov 3-5) and the proceedings on a USB stick. The main conference also includes breakfast, lunch, coffee breaks and a banquet on Nov 4th (Wednesday).
Workshop registration includes access to all workshops and tutorials (Nov 1-2) and a conference USB stick (with workshop and main conference papers). The workshop will also include two coffee breaks daily. Lunch will not be provided for the workshops, however, the workshops are located next to a wide range of affordable food options.
Registration Cancellation Policy Cancellations are required to be made up to 14 days (on/before Oct. 18) in advance. Otherwise, no refund will be issued. A processing fee of SGD$50 applies for all approved refunds.
**On-site On-site registrations must be made by credit card payment only. We cannot accept cash.

Registration Links (and Visa Letter Request)

ACCV 2014 registration is now open. There are three sites for participants to use. The first link is for workshop and conference regitration. If you require an invite letter to obtain a visa to travel to Singapore, please indicate this in the registration portal (there is a check box for those requiring letters). The second link is to pay the publication charges required for all main conference papers (workshops papers do not need to pay this fee). The third link is for those holding complimentary registration codes (e.g. conference sponsors).

Workshop and Full Conference Registration ( link )
Payment of paper publication fee required for all main conference papers ( link )
Complimentary Registration ( link )

If you experience problems with the registration, please contact Mr. Ngee Yong KEE (comkeeyn {AT} nus.edu.sg). We will try to resolve these problems quickly. For those of you in Europe and the US, do be mindful of the time difference (we might not reply for a few hours).

Some people are experiencing problems due to browser cookies. If you need to register and pay using more than one site (e.g. registration and then publication fee), please do it one after another. Do not attempt to open two websites (e.g. registration and publication fees) at the same time. After you have finished with one website, please close the browser entirely before opening another to do your registration and payment.

Payment for the registration and publication fees are made through a 3rd party website (eNets). After you have done your "Workshop and Full Conference Registration" or "Payment of paper publication fee", you should receive one confirmation email for your registration. This email will contain registration or paper publication details. And upon successful payment at eNets, you may print out the eRecipt from the website. Please note that there will be no confirmation email for your payment.

If you have any problems or queries regarding registration, publication fees or visa letter, please contact Ms. Siew Fong Ho as follows:
email : hosf (at) comp.nus.edu.sg
subject : ACCV 2014 Payment or ACCV 2014 Visa Letter


Venue

ACCV2014 will be held on the campus the National University of Singapore. The workshops and tutorials (Nov 1-2, 2014) will be held at the Stephen Riady Centre which is part of the newly constructed University Town. The main conference (Nov 3-5, 2014) will be held at the NUS University Cultural Centre (UCC).
Nov 1-2, Workshops & Tutorials
Stephen Raidy Centre, University Town
National University of Singapore
Nov 3-5, Main Conference
University Cultural Centre
National Univresity of Singapore

Hotels

The following hotels have given ACCV participants special rates for the conference dates. Several will also provide transport in the morning to the conference site.

Hotel Name Rate Morning Transport Breakfast Included Free wifi Additional Comments Booking Info
Bay Hotel S$180++ Yes (and return transport) Yes Yes Complimentary daily mini bar Use form ( pdf )
Park Avenue S$200++ Yes (and return transport) Yes Yes Evening return bus. Use form ( pdf )
Orchard Hotel S$230++ Yes Yes Yes Rate only good before Oct 20th Use link ( click )
Carlton City S$250++ Yes Yes Yes N/A Use form ( pdf )
Hilton Hotel 15% off best rate No - - N/A Use link ( click )


 

Other Hotel Options

Good rates for hotels in Singapore can be found using the following hotel booking sites:
zuji.com
booking.com

Hotel Suggestions Near Downtown Area
4 star Swisshotel Merchant Court, Gallery Hotel Singapore, Studio M Hotel- Singapore
3 star Robertson Quay , Park Hotel Clarke Quay, Fragrance Hotel - Riverside, The Porcelain Hotel, Hotel Re!
2 star Hotel 81 - Chinatown, The Inn at Temple Street, Santa Grand Hotel - Chinatown
Hostels*
Blissful Loft, City Center Backpacker Hostel, Five Stones Hostel, City Backpackers, Racksack Inn (Hongkong Street), Wake Home, Capsule Hotel

*Hostels are the cheapest option and can be found on booking.com (see here ).

Transportation Information

Many of the hotels are providing a bus service to the venue in the morning. For those of you staying in hotels not offering a bus service, we have chartered a shuttle bus service from the closest subway (MRT) station, Kent Ridge Station. Details on how to reach the conference venues (workshops/tutorials and main conference) can be found at the following link (link).

About Singapore

Singapore is a vibrant, cosmopolitan city-state that is home to over 5 million people. Singapore's culture is one of the most diverse in Asia with influences from its mix of Chinese, Indian, Malay, Eurasian and Caucasian population. English is the common language, however, Mandarin, Tamil, and Malay are also spoken. Singapore's Changi airport is consistently ranked among the best airports in the world and can be reached directly from over 200 locations. Singapore has an extensive and affordable public transportation network and world class tourist attractions and dining options.
For more details click here (link).

Sponsors

Sponorship information can be found here ( link )

Platinum Sponsors

Gold Sponsors

 

Silver Sponsors

           

Bronze Sponsors

           

Supporters

Supported by

Held in