[NEW] Please go to our new webpage for WICED 2024 for the most recent information on the competition! (this page will no longer be updated)
[NEW] Upcoming online information session scheduled for 4 September 2023 15:00 BST. Contact us for connection details.
Objectives
The competition aims to inspire and showcase the state of the art in automated editing systems and multimedia processing techniques. In this first edition, we have chosen the BBC old school dataset and offer an aesthetic and a technical challenge that can be done with this dataset.
Materials
- The BBC old school dataset – you’ll need to arrange access to this data in order to enter the competition.
- A description of the dataset presented at WICED 2022 (this is a summary – full details of the dataset contents are provided within the dataset itself).
There are no restrictions on the code, tools or approach used in creating the entries.
Guidelines / Tips
- Start with the example timeline to work with the pre-selected takes – this is provided as an industry-standard EDL file in the dataset (but it’s very human-readable).
- A pre-mixed audio track is also provided in the dataset.
- You can begin with the proxy (lower resolution) video footage if the full-resolution material is inconvenient to download/process
- An example human edit based on the example timeline is also provided.
Judging
There are two entry types
- Aesthetic: trying to obtain the best edit possible
- Technical: exploring some aspect of the dataset without necessarily generating a full edit, such as generating interesting framings
Calendar
- Competition starts: 1 July 2023
- Online information sessions:
- [NEW] 4 September 2023 15:00 BST — contact us for connection details
- Follow up sessions in October – December 2023 to be announced
- Competition ends: 31 January 2024
- Announcement of results: 28 February 2024
- Technical paper deadline: WICED 2024 (TBA)
- Presentation event: WICED 2024 (TBA)
Submission
Entries to the competition should be submitted by online form. The submission should include:
- Author(s), affiliation, and contact
- Download links to:
- the video entry and
- description of the implementation (max 4 pages, A4 PDF)
Authors are invited to also submit a paper to the next WICED (planned for 2024) describing their method.
** Submission link to arrive end of August 2023 **
Tips and resources
- Slides of the competition announcement
- Whitepaper of the dataset
- AutoPod multi-camera editor in Adobe Premier
Stay informed
If you would like to receive updates about the competition, information sessions, and future sessions of WICED, please fill in the form below: