Intelligence Is beyond Learning: A Context-Aware Artificial Intelligent System for Video Understanding

Table 4

Actions recognized by deep learning versus concepts extracted from the context.

Video ID	Concepts extracted from the context	Actions recognized by deep learning
1	0.197 –>video, 0.197 –>parent, 0.105 –>need, 0.076 –>watch, 0.057 –>learn	0.302 –>lecturing, 0.112 –>adult + male + speaking, 0.069 –>pointing, 0.041 –>teaching, 0.036 –>discussing
2	0.311 –> video, 0.278 –>like, 0.198 –>thanksgiving, 0.104 –>psycho, 0.030 –>dinner	0.078 –>dining, 0.051 –>drinking, 0.044 –>discussing, 0.044 –>serving, 0.031 –>autographing
3	0.710 –>scream, 0.137 –>hear, 0.049 –>judge, 0.036 –>murder, 0.036 –>bond	0.135 –>discussing, 0.113 –>pointing, 0.075 –>arresting, 0.065 –>adult + male + speaking, 0.035 –>talking
4	0.801 –>ever, 0.192 –>gmod, 0.004 –>rage, 0.002 –>subscribe	0.202 –>aiming, 0.111 –>mowing, 0.046 –>playing + videogames, 0.040 –>loading, 0.040 –>pointing
5	0.5 –>major, 0.25 –>watch, 0.25 –>video	0.132 –>shopping, 0.041 –>stealing, 0.029 –>buying, 0.027 –>playing, 0.023 –>child + speaking
6	0.081 –>child, 0.06 –>flight, 0.060 -–>attend, 0.057 –>other, 0.056 –>like	0.097 –>waxing, 0.059 –>taping, 0.035 –>cleaning, 0.028 –>pressing, 0.027 –>preaching
7	None	0.126 –>bicycling, 0.062 –>juggling, 0.038 –>running, 0.037 –>feeding, 0.036 –>kicking
8	None	0.053 –>dancing, 0.050 –>adult + male + singing, 0.049 –>playing + videogames, 0.044 –>wrestling, 0.035 –>adult + female + singing
9	0.946 –>scream, 0.036 –>video, 0.0178 –>today	0.061 –>adult + male + singing, 0.047 –>twisting, 0.040 –>racing, 0.031 –>driving, 0.027 –>performing
10	0.376 –>water, 0.261 - > scream, 0.138 –>like, 0.130 –>banshe, 0.029 –>sleep	0.093 –>tickling, 0.053 –>laughing, 0.026 –>giggling, 0.026 –>adult + male + speaking, 0.024 –>juggling