VBS2019 will be part of the International Conference on MultiMedia Modeling 2019 (MMM2019) in Thessaloniki, Greece, and organized as a special side event to the Welcome Reception on January 9, 2019. It will be a moderated session where participants solve Known-Item Search (KIS) and Ad-Hoc Video Search (AVS) tasks that are issued as live presentation of scenes of interest, either as a visual 20 seconds clip (randomly selected), or as a textual description. The goal is to find correct segments (for KIS exactly one segment, for AVS many segments) as fast as possible and submit the segment description (video id and frame number) to the VBS server, which evaluates if the found segment is correct.
TRECVID 2018 Dataset
VBS2019 will collaborate with TRECVID 2018 (i.e. with the Ad-Hoc Video Search (AVS) Task) and use the same IACC.3 dataset, consisting of 4593 Internet Archive videos (144GB, 600 total hours) using videos with durations between 6.5 min and 9.5 min. Please register for TRECVID or contact George Awad at NIST to get access to the IACC.3 data set.
AVS and KIS Tasks
We plan to test at least 20 search tasks:
- 10 AVS tasks, which are a random subset of the 30 AVS tasks of TRECVID 2018. Each AVS task has several/many target shots that should be found.
- 10 KIS tasks, which are selected completely random on site. Each KIS task has only one single 20-seconds long target segment
VBS Server and Testing
The VBS uses an own VBS Server to evaluate found segments for correctness. Therefore, all participants need to connect to the server via a dedicated network (typically Ethernet with CAT-5) and need to submit found segments to the server via a simple HTTP-like protocol. The server is connected to a beamer and presents the current score of all teams in a live manner (in addition to presenting task descriptions). The server as well as example tasks from the previous years are provided here.
Anyone with an exploratory video search tool that allows for retrieval, interactive browsing, exploration in a video collection may participate.
There are no restrictions in terms of allowed features, except for presentation screen recording during the competition, which is disallowed. That means in addition to interactive content search you can use any automatic content search as well.
To apply for participation please submit an extended demo paper (4-6 pages in Springer LNCS format) until the deadline via the MMM 2019 Submission System (please select the “Video Browser Showdown” track). The submission should include a detailed description of the video search tool (including a screenshot of the tool) and show describe how it supports interactive search in video. Submissions will be peer-reviewed to ensure maximum quality. Accepted papers will be published in the proceedings of the MMM conference and should also be presented as poster during the VBS session.
We plan to write a joint journal paper after the VBS competition, where each participating team should contribute to. The winning team will be honored to be in charge of the journal paper (as a main author).