This article presents a novel method for estimating 19 parameters of RGB and depth camera calibration simultaneously. The proposed algorithm is based on applying metaheuristic methods for deep optimization and estimating all parameters of intrinsic, extrinsic, and lens distortions of cameras. This paper compares four metaheuristic algorithms, i.e. a genetic algorithm, particle swarm optimization, the colonial competitive algorithm, and the shuffled frog leaping algorithm, with a numerical algorithm called singular value decomposition. The proposed method does not need the initial estimation for optimization and it can avoid being trapped in local minima. By using nondirect estimation, we achieve middle computing matrices such as the homography matrix, which is used in the pinhole camera model. Both versions of Kinect sensors are used for the experimental evaluation. The mean square of the reprojection error criteria is defined as the objective function in the proposed algorithm. The experimental results show that the proposed method is more efficient and accurate than traditional numerical solutions.