• B.P.Santosh Kumar
• Shafiulla Basha Shaik
• M. Sai Jahnavi
• E.Naveen
• K.Prasannalakshmi, M.Akhila

Depth estimation from a single image is a crucial task in computer vision with applications ranging from autonomous driving to augmented reality. In this project, we employ the UNET++ architecture, a variant of the popular UNET model, for depth estimation. The model is trained on a diverse dataset containing images paired with their corresponding depth maps. During training, we monitor several metrics including training loss, validation loss, structural similarity index (SSIM) scores, and mean squared error (MSE) loss, to ensure both accurate depth predictions and model generalization. Our experiments demonstrate the effectiveness of the UNET++ model in capturing intricate depth details, as indicated by high SSIM scores and low MSE losses on both training and validation sets. This project contributes to advancing depth estimation techniques, offering insights into improving depth perception.

UNET++ architecture, MSE losses, SSIM losses, training losses, validation losses

"EXTRACTING THE FEATURE OF COLOUR IMAGES USING DEPTH ENHANCEMENT TECHNIQUES", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 4, page no.o340-o348, April-2024, Available :http://www.jetir.org/papers/JETIR2404F50.pdf

"EXTRACTING THE FEATURE OF COLOUR IMAGES USING DEPTH ENHANCEMENT TECHNIQUES", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 4, page no. ppo340-o348, April-2024, Available at : http://www.jetir.org/papers/JETIR2404F50.pdf