Improving Few-Shot Image Classification Using Machine- and User-Generated Natural Language Descriptions
This work addresses the problem of improving few-shot learning in computer vision by leveraging language descriptions, offering incremental advancements in model interpretability and performance.
The paper tackled few-shot image classification by using natural language descriptions, showing that their LIDE model with machine-generated descriptions outperformed baselines and improved further with user-generated descriptions, achieving consistent explanations for predictions.
Humans can obtain the knowledge of novel visual concepts from language descriptions, and we thus use the few-shot image classification task to investigate whether a machine learning model can have this capability. Our proposed model, LIDE (Learning from Image and DEscription), has a text decoder to generate the descriptions and a text encoder to obtain the text representations of machine- or user-generated descriptions. We confirmed that LIDE with machine-generated descriptions outperformed baseline models. Moreover, the performance was improved further with high-quality user-generated descriptions. The generated descriptions can be viewed as the explanations of the model's predictions, and we observed that such explanations were consistent with prediction results. We also investigated why the language description improved the few-shot image classification performance by comparing the image representations and the text representations in the feature spaces.