Image colorization provides significant data handling advantages by completely eliminating the need for labeling. This work introduces advancements in the process of converting grayscale images into color using modern machine learning techniques. By utilizing the Lab color space, where luminance (L) is processed separately from the color channels (ab), this article focuses on refining and developing the Conditional Variational Autoencoder (CVAE) model to address the challenges in image colorization with the LFW dataset. Additionally, the Weights & Biases (wandb) tool is employed to support the monitoring of the training process. The training results indicate that the CVAE model achieves commendable outcomes, with a Peak Signal-to-Noise Ratio (PSNR) of 27.63 dB and a Structural Similarity Index Measure (SSIM) of 91.97%. These results highlight the potential of the CVAE model for effective and efficient image colorization.