CHIP: Contrastive Hierarchical Image Pretraining
This work addresses few-shot classification for computer vision applications, but it appears incremental with limited novelty in method and scope.
The paper tackles few-shot object classification by proposing a hierarchical contrastive loss model that classifies unseen objects into general categories, achieving satisfactory results on a custom dataset derived from ImageNet animal classes.
Few-shot object classification is the task of classifying objects in an image with limited number of examples as supervision. We propose a one-shot/few-shot classification model that can classify an object of any unseen class into a relatively general category in an hierarchically based classification. Our model uses a three-level hierarchical contrastive loss based ResNet152 classifier for classifying an object based on its features extracted from Image embedding, not used during the training phase. For our experimentation, we have used a subset of the ImageNet (ILSVRC-12) dataset that contains only the animal classes for training our model and created our own dataset of unseen classes for evaluating our trained model. Our model provides satisfactory results in classifying the unknown objects into a generic category which has been later discussed in greater detail.