Captioning Images on Mobile Devices Using Semi-Statistical Extraction
Ari Ernesto Ortiz Castellanos, Jorge Enrique Román Avalos · 2019
In this work, we propose a semi-statistical Image Extracting Image Semantic Framework using TensorFlow for Mobile Devices. We used datasets for training and generate extraction from other datasets of sentences using the N-Best algorithm selecting the closest sentence associated to images. In addition, we test our experiment with two models of Object Recognition and make comparisons between the different datasets of images and sentences.