Long-term Tracking

The task space of long-term tracking, formed by VOTLT2019 and LaSOT.


Recently, researchers propose long-term tracking to satisfy the demands of real scenarios. However, it is hard to separate the short-term and long-term in the time dimension. Although short-term videos are usually shorter than one minute, only adopting one minute as the task boundary is biased. Therefore, the VOT competition proposes a new criterion -- a task that allows the target to disappear completely can be regarded as long-term tracking. As shown in the video demo, a target may disappear utterly due to out of view or be fully occluded, which is excluded in the short-term tracking environment.

In this work, we select VOTLT2019 and LaSOT to form long-term tracking space.


Will be updated soon.


Will be updated soon.