Few-shot classification studies the problem of quickly adapting a deep learner to understanding novel classes based on few support images. In this context, recent research efforts have been aimed at designing more and more complex classifiers that measure similarities between query and support images, but left the importance of feature embeddings seldom explored. We show that the reliance on sophisticated classifier is not necessary and a simple classifier applied directly to improved feature embeddings can outperform state-of-the-art methods. To this end, we present a new method named textbf{DCAP} in which we investigate how one can improve the quality of embeddings by leveraging textbf{D}ense textbf{C}lassification and textbf{A}ttentive textbf{P}ooling. Specifically, we propose to pre-train a learner on base classes with abundant samples to solve dense classification problem first and then fine-tune the learner on a bunch of randomly sampled few-shot tasks to adapt it to few-shot scenerio or the test time scenerio. We suggest to pool feature maps by applying attentive pooling instead of the widely used global average pooling (GAP) to prepare embeddings for few-shot classification during meta-finetuning. Attentive pooling learns to reweight local descriptors, explaining what the learner is looking for as evidence for decision making. Experiments on two benchmark datasets show the proposed method to be superior in multiple few-shot settings while being simpler and more explainable. Code is available at: url{https://github.com/Ukeyboard/dcap/}.
تحميل البحث