Published Paper Details:

Image captioning is an interesting and challenging task with applications in diverse domains such as image retrieval, organizing and locating images of users' interest, etc. It has huge potential for replacing manual caption generation for images and is especially suitable for large-scale image data. Recently, deep neural network-based methods have achieved great success in the field of computer vision, machine translation, and language generation. In this research, we offer an encoder-decoder based model that can produce image captions with proper grammar. This model uses LSTM as a decoder and VGG16 Hybrid Places 1365 as an encoder. The model is trained on labeled Flickr8k and MS-COCO Captions datasets to guarantee full ground truth accuracy. Further, All widely used standard metrics, including BLEU, METEOR, GLEU, and ROUGE L, are used to evaluate the model. According to experimental findings, the suggested model achieved BLEU-1 score of 0.7350, METEOR score of 0.4768, and GLEU score of 0.2798 on the MS-COCO Caption dataset and BLEU1 score of 0.6666, METEOR score of 0.5060, and GLEU score of 0.2469 on the Flickr8k dataset. Comparing the suggested method to state-of-the-art techniques, a notable improvement in performance was thus obtained. We also present the outcomes of caption generation from real sample photographs, which support the validity of the suggested method and help assess the model's effectiveness even further.