Granular Video Computing: with Rough Sets, Deep Learning and in IOT by Debarati B Chakraborty and Sankar K Pal

Granular Video Computing: with Rough Sets, Deep Learning and in IOT by Debarati B Chakraborty and Sankar K Pal

Author:Debarati B Chakraborty and Sankar K Pal
Language: eng
Format: epub
Publisher: World Scientific Publishing Co. Pte. Ltd.
Published: 2021-04-15T00:00:00+00:00


Table 3.4:Computation time and accuracy of NGrRB, RU and NRBFG.

In case of sequence M_9, the left hand of the person was moving initially, and the right hand starts its movement from 30th frame onwards. As expected, the movement of the right hand cannot be classified by NGrRB (Figs. 3.10.1(a) and 3.10.1(b)) due to the inconsistency between the rules 8 and 9 as in Table 3.2 which results in less coverage (as shown in Table 3.4). However, it is successfully tracked by NRBFG (see Figs. 3.10.1(b) and 3.10.2(b)). The similar situation occurs in case of M_4 sequence, where the right hand was moving initially and it gets stopped at 17th frame, and the left hand starts its movement. Nothing can be tracked by NGrRB (see Figs. 3.10.3(a) and 3.10.4(a)) in this scenario, whereas the moving left hand can be successfully tracked by NRBFG (see Figs. 3.10.3(b) and 3.10.4(b)). It is evident from Table 3.4 that RU gives good results but is more time-consuming as expected, whereas NRBFG keeps a good balance between the time and accuracy.



Download



Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.