Open Access

Table 3

Running time of our method and that of Yang et al. (2019).

Dataset Date Time interval (UT) Frames Average time of our model (s) Average time of Yang et al. (2019) (s)
DS2 2018-06-07 16:17:44–22:32:31 741 0.66 2.06

DS3 2019-06-18 16:34:19–22:14:45 876 0.73 1.97

DS4 2019-06-24 16:15:55–20:55:32 999 0.73 1.99

DS5 2020-07-18 16:41:42–21:20:53 723 0.65 2.09

DS6 2021-04-30 16:38:29–20:59:58 381 0.65 2.11

Notes. The first two columns indicate the dataset and its corresponding date, while the third and fourth columns show the time span and the number of frames in the dataset. The fifth column shows the average time per frame for our proposed method for each time span, and the sixth column shows the average time per frame for the method proposed by Yang et al. (2019). For ease of comparison, we highlight the shorter runtime in bold font for each time span.

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.