Hybrid deep neural network for Bangla automated image descriptor

• DOI: 10.26555/ijain.v6i2.499
• Publication Date: 2020-07-31T14:30:09Z
• Authors: Md Asifuzzaman Ji...; Khan Raqib Mahmud; Abul Kalam Al Azad; Md Shahabub Alam; Anif Minhaz Khan
• Keywords: Closed captioning; Benchmark (surveying)

Automated image to text generation is a computationally challenging computer vision task which requires sufficient comprehension of both syntactic and semantic meaning an generate meaningful description. Until recent times, it has been studied limited scope due the lack visual-descriptor dataset functional models capture intrinsic complexities involving features image. In this study, novel was constructed by generating Bangla textual descriptor from visual input, called Natural Language Image Text (BNLIT), incorporating 100 classes with annotation. A deep neural network-based captioning model proposed The employs Convolutional Neural Network (CNN) classify whole dataset, while Recurrent (RNN) Long Short-Term Memory (LSTM) sequential representation text-based sentences pertinent description based on modular When tested new accomplishes significant enhancement centrality execution for recovery assignment. For experiment that task, we implemented hybrid model, achieved remarkable result self-made Bangladesh perspective. brief, provided benchmark precision in characteristic syntax reconstruction comprehensive numerical analysis results dataset.