Controlling Media Operations On An Automated Local Machine Using Media Operations

Computer Science Engineering

In everyday life, people speak with one another and utilize an expansive scope of motions during the time spent connection.Aside from relational correspondence, numerous hours are gone through in the association with electronic gadgets. Hand signal acknowledgment is an arduous assignment to tackle in recordings. In this paper, we use a residual attention network that has been trained from start to finish to recognise hand motions. Given the numerous attention blocks stacked on top of each other, a network is fabricated which makes various highlights at every consideration block. Our consideration based residual network (ResATN) can be fabricated and stretched out to extremely profound layers. Utilizing this network, a broad examination is performed on different networks dependent on three freely accessible datasets. We evaluated our network using one of the finest recognition dataset(Jester) and achieved competitive performances by training the network from scratch. This hand signal recognition system will not only replace the use of a mouse to monitor the media player, but will also include various gesture commands that will be useful in controlling the programme.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.