LikeLines is an enhanced video player that provides users with a navigable heat-map of interesting regions below the player for the videos they are watching. This so called "like line" reflects what users liked and which parts they have watched or skipped. The "like line" heat-map can be initially seeded through content analysis methods to address the cold start problem.
This SMILA component provides access to the data that is collected by a LikeLines back-end server. The pipelet provided by this component computes the most interesting time-points for a given video based on the information it requested from a configured LikeLines server. If no sufficient information is available to compute the most interesting parts of a video, the pipelet can start an indexing service to perform content analysis on the given video as a substitute.
The source code of the LikeLines SMILA component is maintained at:
The LikeLines project page is hosted at:
The LikeLines system is being developed under the CUbRIK project: