Multimodal Objects Categorization by Fusing GMM and Multi-layer Perceptron
Aysha Naseer, Ahmad Jalal
Abstract
Multi-object recognition resolves the need to instantaneously identify numerous things, which is essential for developing an in-depth understanding of the world of vision. Variations in object features, occlusions, scale, perspective, and intricate backgrounds make this task challenging. The practical strategy for categorizing objects into multiple categories using well-known benchmark datasets is presented in this work. The method extracts multi-object segmentation from difficult images using an advanced Gaussian mixture model (GMM) technique. A multi-layer perceptron (MLP) architecture that incorporates local descriptors and detectors optimizes the classification process. The method’s ability to handle concerns like object occlusion and spatial complexity is validated by extensive experiments using a well-known dataset, Corel 10k. The advised strategy overtakes cutting-edge techniques in terms of accuracy, with scores of 87.40% accuracy for the Corel 10k dataset.