Articulated Human Motion Tracking Using Sequential Immune Genetic Algorithm

Algorithm 2

IGA-based pose optimization algorithm.

Input: total number of antibodies 𝑁, maximum number of generations 𝐾.

Steps:

 (1) Initialization: generate the initial population with 𝑁 antibodies. The initial population can

  be represented as 𝐴(0)={𝑥₁(0),𝑥₂(0),…,𝑥_𝑁(0)}.

 (2) Repeat:

   While  (𝑘<𝐾)  do

   (2.1) Genetic operator: Perform genetic operators 𝑇^𝑝_𝑔 on the 𝑘th parent 𝐴(𝑘) and obtain the

   results 𝐵(𝑘);

   (2.2) Vaccination: Perform vaccination 𝑇^𝑝_𝑣 on 𝐵(𝑘) and obtain 𝐶(𝑘);

   (2.3) Immune selection: Perform immune selection 𝑇^𝑝_𝑠 on 𝐶(𝑘) and obtain the next generation

   population 𝐴(𝑘+1);

   (2.4) 𝑘=𝑘+1.

   End while

 (3) Output: output the population 𝐴(𝐾)={𝑥₁(𝐾),𝑥₂(𝐾),…,𝑥_𝑁(𝐾)}.