Caption Generation for Images with Deep Neural Networks
Anusha Anil, S. Santhanalakshmi · 2022 2nd International Conference on Intelligent Technologies (CONIT) · 2022
In an evolving environment of technologies, Deep Learning and the models have enabled the implementation of complex problem statements. Human beings learn to observe and detect, and produce or express descriptions of them and as they grow, can do so for the various aspects in their surrounding and how they interact with each other. Caption generation for images is one such task as it involves the extraction of features and then also further seeing the interactions amongst the recognised objects/aspects and producing descriptions of what and how it is. The various CNN models and how they contribute to the problem statement has been an interesting point of motivation and that along with other architectures are to be implemented for the purpose of caption generation.