为了解决这个问题,Yale Song以3秒为一段将视频截开,在各段之间插入时间间隔。
That allowed the computer time for reflection.
这样就给计算机留出了响应时间。
Its accuracy was also increased by interpreting each block in light of those immediatelybefore and after it,
同时识别的准确率也有所提高。因为这样计算机就能根据前一段及后一段的视频来理解当前这一段,
to see if the result was a coherent message of the sort a deck officer might actually wish toimpart.
看看结果是不是那种甲板指挥员可能真的想摆出的有特定意义的手势。
The result is a system that gets it right three-quarters of the time.
他们最终做出了一个正确率为75%的系统。
Obviously that is not enough:
显然,那样是不够的:
you would not entrust the fate of a multi-million-dollar drone to such a system.
你不会将一架价值数百万美元的无人机交给这样的系统。
But it is a good start.
但这是个好的开始。
If Mr Song can push the accuracy up to that displayed by a human pilot, then the task ofcontrolling activity on deck should become a lot easier.
【2015考研英语阅读无人机】相关文章:
最新
2016-10-18
2016-10-11
2016-10-11
2016-10-08
2016-09-30
2016-09-30