Multi-level Similarity Learning for Low-Shot Recognition
This addresses the problem of recognizing unseen objects with very few examples for computer vision applications, but it appears incremental as it builds on existing similarity learning methods.
The paper tackles low-shot recognition by proposing a multi-level similarity model (MLSM) to classify unseen objects with limited labeled samples, achieving promising results in 5-way experiments on Caltech-UCSD datasets.
Low-shot learning indicates the ability to recognize unseen objects based on very limited labeled training samples, which simulates human visual intelligence. According to this concept, we propose a multi-level similarity model (MLSM) to capture the deep encoded distance metric between the support and query samples. Our approach is achieved based on the fact that the image similarity learning can be decomposed into image-level, global-level, and object-level. Once the similarity function is established, MLSM will be able to classify images for unseen classes by computing the similarity scores between a limited number of labeled samples and the target images. Furthermore, we conduct 5-way experiments with both 1-shot and 5-shot setting on Caltech-UCSD datasets. It is demonstrated that the proposed model can achieve promising results compared with the existing methods in practical applications.