Comparative Analysis of Deep Learning Models for Facial Emotional Recognition Kushagra Srivastava, Sanjeev Sharma 2024 11th International Conference on Reliability Infocom Technologies and Optimization Trends and Future Directions Icrito 2024, 2024 Emotional Intelligence and Facial Emotion Recognition are some of the fields that come under high focus when computer vision is finding its way to develop one of the smartest AI solutions. Implementation of custom models often requires elaborate and well-defined processes at all stages of model development i.e. dataset preparation, model training, tuning, and optimization to run them effectively in applications. Thus, the use of pre-trained models comes into play, which can significantly reduce the training/tuning cost and time requirements. Thus, analysis of publicly available pre-trained models contributes significantly to their use for building efficient and faster AI solutions.
Acne Detection Care System using Deep Learning Rohit Yadav, Aashika Jain, Sanjiv Sharma 2024 11th International Conference on Reliability Infocom Technologies and Optimization Trends and Future Directions Icrito 2024, 2024 Millions of people worldwide suffer from acne, a common dermatological ailment that frequently causes both physical and psychological discomfort. The prevalence of acne, a common skin condition, poses a significant challenge to derma-tologists and individuals seeking effective skincare solutions. This research introduces ‘Acne Care’, an innovative system that leverages deep learning techniques and Reset18 application for the detection and personalized care of acne. This model analyse various skin abnormalities and make a severity detection system based on the classification using deep learning algorithms. This ensemble model could accurately predict the number, location, and severity of acne at the same time. It might also be a useful tool for the patient to self-test and help the doctor diagnose them. This paper presents the development, methodology, and potential impact of this model, addressing the growing need for more efficient and effective acne management. The findings of this research paper contribute to the development and advancement of deep learning based regression models to assess the severity level of acne lesions from selfie images and their management.
Speech Emotion Recognition using Deep Learning Divya Garg, Amrita Singh, Anshika Gupta, Sanjiv Sharma 2024 11th International Conference on Reliability Infocom Technologies and Optimization Trends and Future Directions Icrito 2024, 2024 Speech Emotion Recognition (SER) is an emerging field that involves recognizing emotions conveyed in speech. Emotions expressed through speech can greatly impact decision-making. This paper delves into the topic of speech emotion recognition (SER) and its focus on interpreting emotions conveyed through spoken language. The importance of SER lies in its potential to improve human-computer interaction, cognitive analysis, and psychiatric assessment. The study combines and preprocesses audio data from various datasets, such as RAVDESS, CREMA-D, TESS, and SAVEE, and uses log mel spectrograms to effectively extract features. Various methods including CNN models, and standard and optimized feature extraction techniques are used. The results suggest that SER has significant real-world applications and the approaches provided effectively identify emotional and voice signals.
A Comprehensive Study of the Machine Learning with Federated Learning Approach for Predicting Heart Disease Pakhi Sharma, Sanjiv Sharma Proceedings of International Conference on Contemporary Computing and Informatics Ic3i 2023, 2023 Heart disease is a leading cause of mortality worldwide, resulting in millions of deaths annually. As individuals age and their physical condition deteriorates, the risk of developing heart disease increases. To mitigate this risk, predictive models leveraging machine learning and artificial intelligence have emerged as valuable tools for early diagnosis and treatment. In this review paper, we introduce the Google-pioneered concept of federated learning as a means to address concerns about data safety in the context of heart disease prediction. Federated learning, also known as collaborative learning, employs a technique wherein an algorithm is trained through multiple independent sessions, each utilizing its own dataset. This paper aims to provide a comprehensive investigation of recent machine learning approaches and databases employed in predicting the occurrence of cardiovascular disease.
