Abstract: Manual colorization of black and white pictures could be a difficult errand and wasteful. It has been endeavored utilizing Photoshop editing, but it demonstrates to be troublesome because it requires broad investigate and a picture can take up to one month to colorize. A practical approach to the assignment is to actualize advanced picture colorization methods. The literature on picture colorization has been an range of intrigued within the final decade, because it stands at the juncture of two arcane disciplines, advanced picture preparing and profound learning. Endeavors have been made to utilize the ever-increasing availability of end-to-end profound learning models and use the benefits of exchange learning. Picture features can be consequently extricated from the preparing information utilizing profound learning models such as Convolutional Neural Systems (CNN). This could be assisted by human mediation and by utilizing as of late created Generative Antagonistic Systems (GAN). We actualize picture colorization utilizing different CNN and GAN models whereas leveraging pre-trained models for way better highlight extraction and compare the execution of these models.
Key Words: Deep learning, Pre-trained model, CNN, GAN, image colorization, Pix2pix
Cite:
Dr.S. Govindaraju, Gowtham T, "COLORIZATION OF BLACK AND WHITE IMAGES", IJARCCE International Journal of Advanced Research in Computer and Communication Engineering, vol. 13, no. 3, 2024, Crossref https://doi.org/10.17148/IJARCCE.2024.133141.
|
DOI:
10.17148/IJARCCE.2024.133141
