Sound Can Help Us See More Clearly

. 2022 Jan 13;22(2):599. doi: 10.3390/s22020599

Algorithm 1 Micro Attention branch Algorithm

Require:

S o u n d W a v e

: Sound waveform data extracted from video;
Ensure:

S o u n d T e x t u r e

1:
for each $n \in [1, 20]$ do
2:
$C h a n n e l_{n}$ = Hilbert(BandPassFiltern( $S o u n d W a v e$ ));
3:
$C h a n n e l_{n}$ = $C h a n n e l_{n}$ **0.3;
4:
$C h a n n e l_{n}$ = Compress( $C h a n n e l_{n}$ );
5:
$M a r g i n a l_{n}$ = Marginal( $C h a n n e l_{n}$ );
6:
$M a r g i n a l D a t a$ = Append( $M a r g i n a l D a t a$ , $M a r g i n a l_{n}$ )
7:
$C o r r e l a t i o n C_{n}$ = Correlation( $C h a n n e l_{n}$ , $C h a n n e l_{20 - n}$ )
8:
$C o r r e l a t i o n D a t e$ = Append( $C o r r e l a t i o n D a t e$ , $C o r r e l a t i o n C_{n}$ )
9:
for each $m \in [1, 10]$ do
10:
$C h a n n e l_{n m}$ = BandPassFilterm( $C h a n n e l_{n}$ );
11:
$M a r g i n a l_{n m}$ = Marginal( $C h a n n e l_{n m}$ );
12:
$M a r g i n a l D a t a$ = Append( $M a r g i n a l D a t a$ , $M a r g i n a l_{n m}$ )
13:
$C o r r e l a t i o n C 1_{m}$ = Correlation( $C h a n n e l_{n} m$ , $C h a n n e l_{n (10 - m)}$ )
14:
$C o r r e l a t i o n C 2_{m}$ = Correlation( $C h a n n e l_{n} m$ , $C h a n n e l_{(20 - n) m}$ )
15:
$C o r r e l a t i o n D a t e$ = Append( $C o r r e l a t i o n D a t e$ , $C o r r e l a t i o n C 1_{m}$ , $C o r r e l a t i o n C 2_{m}$ )
16:
end for
17:
end for
18:
$S o u n d T e x t u r e$ = Append( $M a r g i n a l D a t a$ , $C o r r e l a t i o n D a t e$ )
19:
return $S o u n d T e x t u r e$