|
The main goal of this workshop is to explore scientific research on Big Data in 3D computer vision. There is an increasingly rich amount of 3D visual data available from our daily life.
- Corporations including Google, Microsoft and Nokia have 3D data for major cities in their map services
- Thanks to the SfM technique, photos and videos taken by consumer smartphones amount to a dynamic influx of 3D content which covers our everyday life both globally and instantly.
- Inexpensive 3D sensors such as Kinect pave the way for capturing rich 3D data from people’s entertainments.
This workshop focuses on large scale 3D data analysis algorithms, methods and solutions. It is expected to demonstrate the state-of-the-arts in the convergence of 3D sensor technology, 3D computer vision, 3D visualization and 3D applications. This workshop will cover major aspects of big data vision and will be the venue for papers to highlight the recent advanced research from academic and industry labs through the connection of big data in 3D computer vision.
This workshop aims to foster research efforts related, but not restricted, to following topics:
- Object description, detection, and recognition on large scale point cloud data
- 3D scene understanding, object-based representation and segmentation
- Multiview and 3D video coding
- Matching and registration across point cloud data of various sensing
- Navigation, localization, SLAM and semantic mapping
- 3D motion analysis methods
- Structure from motion and multi-view stereo techniques towards large scale environments
- Benchmark 'Big' 3D datasets
- 3D tracking and reconstruction from live video
- Indexing, searching and alignment of large-scale 3D data
- Content-based 3D retrieval and recognition
- 3D mesh, texture, point, and volume-based representation, 3D scene browsing on mobile devices, 3D electronic-map
- 3D motion animation
- Augmented reality and mixing of virtual and real worlds, augmented reality in street view, web-based 3D map applications.
 Authors are invited to submit a full paper according to the guidelines available on the ACCV2014 conference website. Submissions must be a maximum of 14 pages (not including references). You may include as many additional pages for references as you need. Papers of more than 14 pages of main text will be rejected without review. Submitted papers will be reviewed by the workshop program committee in a double blind manner. Supplementary material can be submitted which may contain illustrative videos, additional results, detailed theorem proofs or material that is under review elsewhere but necessary for understanding the submission (this must be declared as such). Submissions, including supplementary material, must be self-contained. It is not acceptable to include (even anonymised) URLs to additional external material. In submitting a manuscript to this workshop, the authors acknowledge that no paper substantially similar in content has been submitted to another conference or workshop during the review period. Papers violating the formatting rules, the double-blind or dual-submission policies will be rejected without review. The conference proceedings will be published by Springer in the Lecture Notes in Computer Science (LNCS) series. Accepted papers must be registered and presented; otherwise they will not be included in the workshop proceedings.
Special announcement from ACCV2014 Organizing Committee on main conference and workshop paper submission
ACCV2014 will continue the "fall back" strategy used successfully in ACCV2012, which can connect some rejected main conference papers to appropriate workshops. The papers rejected by the main conference but connected to a workshop by the author(s) will be automatically passed to the workshop organizers, together with the main conference review comments. The workshop organizers will see the paper and the reviews, based on which they can decide whether the paper can be accepted or not by the workshop.
For this purpose, when potential ACCV2014 authors submit a paper to the main conference, they will be allowed to select a workshop as a “backup” option in case that the paper is not accepted by the main conference.
Contact
Jian.Zhang@uts.edu.au |
