Artist Based Video Recommendation System using Machine Learning
Sandeep Sukumar Hipparagi, O.V. Ramana Murthy · 2021 Third International Conference on Inventive Research in Computing Applications (ICIRCA) · 2021
This study investigates the use of automatically derived visual characteristics in recommender systems and offers several novel contributions to the field of video recommendations. Because it is difficult for customers to see all of the videos in order to identify the ones they want, or even manually manage these large video collections. The Recommender System is one such engine, which suggests previously undiscovered material that the user may enjoy and assist them in making decisions. As a result, depending on the user's unique artist, id, alignment, identifying face edges, and extracting various characteristics from faces. The algorithm offers personalized sets of videos to users by recognizing human faces in low lighting condition, where the presence of noise in images, scales, masquerade in a diverse environment, and the discovery of unexplored artist videos remains a major obstacle that this work aims to resolve. In recent years, Deep Neural Network (DNN) has been demonstrated for a trained face detector which outperforms end-to-end long-standing approaches. The Support Vector Machine (SVM) is a key component in a range of applications, which includes image recognition, pattern recognition, and classification. One of the most promising areas of research in computer vision is face identification of different patterns. An approach to face recognition using deep learning techniques and an SVM model for face image identification is created in this work, which uses a DNN to extract features and SVM classification methods to identify the artist from videos.