My task is to implement a prototype of a visual object detection system. This study proposes a deep neural network architecture for the phone location detection model. Initially multiple images have been trained using VGG16 Architecture , a phone is considered to be detected correctly on a test image if the final object coordinates of the test image is within a radius of 0.05 (normalized distance) centered on the phone. i have developed visual onject detection model to find a location of a phone dropped on the floor from a single RGB camera image. Accuracy of model observed is 70% of test image split(90 :10)
