Semantic interdisciplinary evaluation of image captioning models

Sirisha, Uddagiri; Chandana, Bolem Sai

doi:10.1080/23311916.2022.2104333

Cited by 13 publications

(9 citation statements)

References 68 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…According to Fig. 8, anger (1,00,445) is the biggest feeling on Twitter associated with the Ukraine-Russia war, followed by optimism (39,284), joy (25,091), and sadness (25,795). Fig.…”

Section: B Experimental Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Aspect based Sentiment & Emotion Analysis with ROBERTa, LSTM

Sirisha¹,

Chandana²

2022

IJACSA

View full text Add to dashboard Cite

Internet usage has increased social media over the past few years, significantly impacting public opinion on online social networks. Nowadays, these websites are considered the most appropriate place to express feelings and opinions. The popular social media site Twitter offers valuable insight into people's thoughts. Throughout the conflict between Russia and Ukraine, people from all over the world have expressed their opinions. In this study, "machine-learning" and "deep-learning" techniques are used to understand people's emotions and their views about this war are revealed. This study unveils a novel deeplearning approach that merges the best features of the sequence and transformer models while fixing their respective flaws. The model combines Roberta with ABSA (Aspect based sentiment analysis) and Long Short-Term Memory for sentiment analysis. A large dataset of geographically tagged tweets related to the Ukraine-Russia war was collected from Twitter. We analyzed this dataset using the Roberta-based sentiment model. In contrast, the Long Short-Term Memory model can effectively capture longdistance contextual semantics. The Robustly optimized BERT with ABSA approach maps words into a compact, meaningful word embedding space. The accuracy of the suggested hybrid model is 94.7%, which is higher than the accuracy of the stateof-the-art techniques.

show abstract

“…According to Fig. 8, anger (1,00,445) is the biggest feeling on Twitter associated with the Ukraine-Russia war, followed by optimism (39,284), joy (25,091), and sadness (25,795). Fig.…”

Section: B Experimental Resultsmentioning

confidence: 99%

“…Deep-learning is used in various aspects such as objectdetection, image-captioning [39], image segmentation [40], etc. This study uses machine and deep learning approaches to assess how twitter discussions regarding the Ukraine-Russia war affect public sentiment.…”

Section: Introductionmentioning

confidence: 99%

Aspect based Sentiment & Emotion Analysis with ROBERTa, LSTM

Sirisha¹,

Chandana²

2022

IJACSA

View full text Add to dashboard Cite

show abstract

“…(4) Deep Learning Based Methods: The development of deep learning has contributed to the development of CNNs that can detect irregular texts [7][8][9][10]. These models are trained on annotated datasets to learn the complex patterns and characteristics of irregular text, enabling them to accurately detect and localize such text in images.…”

Section: Related Workmentioning

confidence: 99%

Geometric Deep Learning for Enhancing Irregular Scene Text Detection

Aluri,

Tatavarthi

2024

RIA

View full text Add to dashboard Cite

Text detection in natural scene images presents significant challenges, particularly in detecting irregular shapes. As a result of the limited receptive field of CNNs, existing methods have difficulty capturing long-range relationships between distant component regions. This study introduces an innovative method for identifying irregular text in images of natural scenes. The approach utilizes a U-net architecture combined with connected component analysis, resulting in improved accuracy in detecting text components and reducing the identification of non-character text components. Additionally, our strategy incorporates the use of graph convolution networks (GCN) to deduce adjacency relations among text components. The integration of GCNs introduces a sophisticated mechanism for inferring adjacency relations, contributing significantly to the advancement of text detection in natural scene images. Our method's efficacy is showcased through experimental assessments on three publicly available datasets: "ICDAR2013," "CTW-1500," and "MSRA-TD500."

show abstract

“…An electronic photograph is a binary representation of visual information. Semantic segmentation, object detection, fake image identification [ 26 ], and image captioning [ 27 ] are just a few examples of areas where convolutional neural networks (CNNs) have seen significant advancements in recent years thanks to the explosion of deep learning. With a CNN-LSTM model, features are extracted from input data using CNN layers, while sequence prediction is accomplished using LSTM layers.…”

Section: The Proposed Modelmentioning

confidence: 99%

Privacy Preserving Image Encryption with Optimal Deep Transfer Learning Based Accident Severity Classification Model

Sirisha

Chandana

2023

Sensors

Self Cite

View full text Add to dashboard Cite

Effective accident management acts as a vital part of emergency and traffic control systems. In such systems, accident data can be collected from different sources (unmanned aerial vehicles, surveillance cameras, on-site people, etc.) and images are considered a major source. Accident site photos and measurements are the most important evidence. Attackers will steal data and breach personal privacy, causing untold costs. The massive number of images commonly employed poses a significant challenge to privacy preservation, and image encryption can be used to accomplish cloud storage and secure image transmission. Automated severity estimation using deep-learning (DL) models becomes essential for effective accident management. Therefore, this article presents a novel Privacy Preserving Image Encryption with Optimal Deep-Learning-based Accident Severity Classification (PPIE-ODLASC) method. The primary objective of the PPIE-ODLASC algorithm is to securely transmit the accident images and classify accident severity into different levels. In the presented PPIE-ODLASC technique, two major processes are involved, namely encryption and severity classification (i.e., high, medium, low, and normal). For accident image encryption, the multi-key homomorphic encryption (MKHE) technique with lion swarm optimization (LSO)-based optimal key generation procedure is involved. In addition, the PPIE-ODLASC approach involves YOLO-v5 object detector to identify the region of interest (ROI) in the accident images. Moreover, the accident severity classification module encompasses Xception feature extractor, bidirectional gated recurrent unit (BiGRU) classification, and Bayesian optimization (BO)-based hyperparameter tuning. The experimental validation of the proposed PPIE-ODLASC algorithm is tested utilizing accident images and the outcomes are examined in terms of many measures. The comparative examination revealed that the PPIE-ODLASC technique showed an enhanced performance of 57.68 dB over other existing models.

show abstract

Semantic interdisciplinary evaluation of image captioning models

Cited by 13 publications

References 68 publications

Aspect based Sentiment & Emotion Analysis with ROBERTa, LSTM

Aspect based Sentiment & Emotion Analysis with ROBERTa, LSTM

Geometric Deep Learning for Enhancing Irregular Scene Text Detection

Privacy Preserving Image Encryption with Optimal Deep Transfer Learning Based Accident Severity Classification Model

Contact Info

Product

Resources

About