Banca de DEFESA: JOAO MEDRADO GONDIM

Uma banca de DEFESA de MESTRADO foi cadastrada pelo programa.
STUDENT : JOAO MEDRADO GONDIM
DATE: 23/08/2023
TIME: 09:00
LOCAL: https://conferenciaweb.rnp.br/ufba/defesa-de-mestrado-do-formas
TITLE:

 INCREASING THE IMAGE CAPTIONING MODELS IN PORTUGUESE THROUGH LINGUISTIC INFORMATION

KEY WORDS:

Image Captioning; Neural Networks; Computer Vision; Natural Language Processing.

PAGES: 93
BIG AREA: Ciências Exatas e da Terra
AREA: Ciência da Computação
SUBÁREA: Metodologia e Técnicas da Computação
SPECIALTY: Banco de Dados
SUMMARY:

The increase in the number of applications that require accessibility, information retrieval and human-computer interaction has culminated in a growing need for automated generation of the description of an image. This automated description requires an identification of the scenario, characters and objects present and how these elements relate to each other. From these elements it becomes possible to generate a sentence in natural language describing the content of the image. The development of methods capable of automatically generating the sentences describing the image permeates a research area called textit{Image Captioning}. Most researches and datasets in the area of Image Captioning focus on the English language, developing models and building efficient state-of-the-art resources. Languages with few resources for development, such as Portuguese, require more research to achieve a descriptive and understandable sentence. However, only the agglomeration of several objects in the image does not generate a sentence in the Portuguese language. In this context, this work proposes the analysis and incorporation of linguistic resources that can guide the language model in generating a description that is more representative of the image and the sentence in Portuguese. Experiments were performed with the translation of datasets to generate the description in Portuguese. The results indicate that the morphological analysis of the outputs of an Image Captioning model, as well as the incorporation of grammatical classes during the training, will contribute to a better description of the image in Portuguese.

COMMITTEE MEMBERS:
Presidente - 1232218 - DANIELA BARREIRO CLARO
Interna - 2115505 - TATIANE NOGUEIRA RIOS
Externa à Instituição - SANDRA ELIZA FONTES DE AVILA - UNICAMP