An Application of Classifier Combination Methods in Hand Gesture Recognition

Algorithm 1

The general algorithmic framework for bagging, random subspace, and random forest methods.

● Input

  A training set ℒ={(𝐱_𝑖,𝑦_𝑖)}^𝑁𝑖=1; A base learner 𝒲; Number of iterations 𝑇; A new data point 𝐱

  to be classified.

● Training Phase

  For 𝑡=1,…,𝑇

   (1) Utilize the corresponding technique (i.e., bootstrap sampling or randomly selecting

   features) to get a training set ℒ_𝑡.

   (2) Provide ℒ_𝑡 as the input of 𝒲 (random forest has an additional randomness injection  

       operation) to train base classifier 𝐶_𝑡.

  EndFor

● Output

   – The class label for 𝐱 predicted by the ensemble classifier 𝐶^∗ as

             𝐶^∗(𝐱)=argmax_{𝑇𝑦∈Φ}∑_𝑡=1𝐼(𝐶_𝑡(𝐱)=𝑦),

   where 𝐼(⋅) denotes the indicator function which takes value 1 or 0 depending on

   whether the condition of it is true or false.