Fine-grained Classification via Categorical Memory Networks
This work provides an incremental improvement in fine-grained classification accuracy for computer vision researchers.
This paper addresses fine-grained classification by introducing a class-specific memory module that stores prototypical feature representations for each category. The method significantly improves accuracy over baseline CNNs and achieves competitive accuracy with state-of-the-art methods on four benchmarks.
Motivated by the desire to exploit patterns shared across classes, we present a simple yet effective class-specific memory module for fine-grained feature learning. The memory module stores the prototypical feature representation for each category as a moving average. We hypothesize that the combination of similarities with respect to each category is itself a useful discriminative cue. To detect these similarities, we use attention as a querying mechanism. The attention scores with respect to each class prototype are used as weights to combine prototypes via weighted sum, producing a uniquely tailored response feature representation for a given input. The original and response features are combined to produce an augmented feature for classification. We integrate our class-specific memory module into a standard convolutional neural network, yielding a Categorical Memory Network. Our memory module significantly improves accuracy over baseline CNNs, achieving competitive accuracy with state-of-the-art methods on four benchmarks, including CUB-200-2011, Stanford Cars, FGVC Aircraft, and NABirds.