Performance of Gaussian Mixture Model Classifiers on Embedded Feature Spaces
This work provides an incremental improvement for researchers in multimedia analysis by optimizing GMM-based classification on pre-trained embeddings.
The study evaluated Gaussian Mixture Model (GMM) classifiers on CLIP and ImageBind embeddings for multimedia classification, finding that one Gaussian component per class often suffices and ImageBind generally outperforms CLIP, even with PCA compression.
Data embeddings with CLIP and ImageBind provide powerful features for the analysis of multimedia and/or multimodal data. We assess their performance here for classification using a Gaussian Mixture models (GMMs) based layer as an alternative to the standard Softmax layer. GMMs based classifiers have recently been shown to have interesting performances as part of deep learning pipelines trained end-to-end. Our first contribution is to investigate GMM based classification performance taking advantage of the embedded spaces CLIP and ImageBind. Our second contribution is in proposing our own GMM based classifier with a lower parameters count than previously proposed. Our findings are, that in most cases, on these tested embedded spaces, one gaussian component in the GMMs is often enough for capturing each class, and we hypothesize that this may be due to the contrastive loss used for training these embedded spaces that naturally concentrates features together for each class. We also observed that ImageBind often provides better performance than CLIP for classification of image datasets even when these embedded spaces are compressed using PCA.