Algorithm 1: Pseudo-code of pig behavior analysis.

Input: A video sequence $V$; object detector $Det$; tracking score threshold $\eta$ is set 0.75; Frames per second $Fps$; Output: Tracks $T$ of the video

1Initialization: $T\leftarrow \varnothing$ 2for frame $f_k$ in $V$ do 3 $D\leftarrow Det\left(f_k\right)$ 4 Initialize time-count array including four elements for time statistics $a\leftarrow \left[0,0,0,0\right]$ 5 Set variable $category\_index\leftarrow D\left\{category\right\}$ 6 $a\left[category\_index\right]\leftarrow 1$ 7 Associate $T$ with $D$ using OC-SORT: 8  if succeed to match then 9   Call the update or re-activate function to update the status of tracks 10   Set variable $A\leftarrow T\left[category\_time\_array\right]+a$ 11   $T\left[category\_time\_array\right]\leftarrow A$ 12  end 13  if $D$ failed to match and $D>\eta$ then 14   Call the function to create a new track. 15   Initialize time-count array $A\leftarrow \left[0,0,0,0\right]$ 16   $T\left[category\_time\_array\right]\leftarrow A$ 17  end 18End 19$T\left[category\_time\_array\right]=T\left[category\_time\_array\right]/Fps$

Return $T$