Using Contrastive Learning and Pseudolabels to learn representations for Retail Product Image Classification
This addresses the challenge of limited data variations in retail product images for classification tasks, though it is incremental as it builds on existing methods like contrastive learning.
The paper tackled the problem of few-shot retail product image classification by pretraining a convolutional neural network backbone using contrastive learning and pseudolabel-based noisy student training, achieving accuracy comparable to full fine-tuning.
Retail product Image classification problems are often few shot classification problems, given retail product classes cannot have the type of variations across images like a cat or dog or tree could have. Previous works have shown different methods to finetune Convolutional Neural Networks to achieve better classification accuracy on such datasets. In this work, we try to address the problem statement : Can we pretrain a Convolutional Neural Network backbone which yields good enough representations for retail product images, so that training a simple logistic regression on these representations gives us good classifiers ? We use contrastive learning and pseudolabel based noisy student training to learn representations that get accuracy in order of finetuning the entire Convnet backbone for retail product image classification.