Fine Grained Class

Fine-grained classification focuses on distinguishing between highly similar sub-categories within a broader class, a challenge amplified by subtle visual differences and limited labeled data. Current research emphasizes developing unsupervised or semi-supervised methods, often employing contrastive learning, hierarchical architectures, and proxy-based learning to leverage coarse-grained labels or limited annotations for improved fine-grained categorization. These advancements are crucial for various applications, including medical image analysis, disaster response, and efficient named entity recognition, where obtaining exhaustive fine-grained annotations is impractical or costly. The development of robust and efficient methods for fine-grained classification is driving progress in numerous fields reliant on accurate and detailed categorization.