MSeeker1340/Vision2018-RL

Final course project for CSCI-GA 3033: Vision meets machine learning

Vision2018-RL

Final course project for CSCI-GA 3033: Vision meets machine learning Instructions: https://d1b10bmlvqabco.cloudfront.net/attach/jm0yvg6k8c072i/j7ay8cyggyt4yk/jp7lj9s8p2ew/Reinforcement_Learning.11232018.pdf

Homework by 12/6

Task 1: Skeleton coordinate transformation

Input:

coordinate(v_i) 

Output:

(X, Y, Z)_i

Task 2: Compute action

Input:

S^t ={s_i^t}, i = 1 to 25
  s_i^t = (v_i, d_i)^t
      where v_i^t=(X_i, Y_i, Z_i)^t
            d_i^t = v_i^{t} - v_i^{t-1}

Output:

A^t ={a_i^t}, i = 1 to 25
  a_i^t = (a_i^x, a_i^y, a_i^z)^t
      where a_i^t = d_i^{t+1}- d_i^{t}