Investigating Deep learning models for NFT classification : A Review
Ashutosh Singla, Mandeep Gupta
Abstract
Present research work is focused on NFT image classification using deep learning techniques. The classification task is extended by incorporating noise removal techniques in addition to image compression before employing the CNN model. After collecting and splitting the dataset, the images undergo both compression and noise removal processes to enhance their visual quality and reduce unwanted artifacts. The preprocessed images are then fed into a CNN model for training and evaluation. This approach aims to improve the model's ability to discern features in images by minimizing the impact of noise. The evaluation phase involves assessing the model's performance on the test set to determine its effectiveness in classifying NFT images after both compression and noise removal.