Subset Feature Learning for Fine-Grained Category Classification
This addresses the problem of fine-grained visual recognition for researchers and practitioners, offering an incremental improvement over existing methods.
The paper tackles fine-grained category classification by clustering visually similar classes and learning subset-specific deep features, achieving a mean accuracy of 77.5% on the Caltech-UCSD bird dataset without bounding boxes at test time, compared to a previous best of 73.2%.
Fine-grained categorisation has been a challenging problem due to small inter-class variation, large intra-class variation and low number of training images. We propose a learning system which first clusters visually similar classes and then learns deep convolutional neural network features specific to each subset. Experiments on the popular fine-grained Caltech-UCSD bird dataset show that the proposed method outperforms recent fine-grained categorisation methods under the most difficult setting: no bounding boxes are presented at test time. It achieves a mean accuracy of 77.5%, compared to the previous best performance of 73.2%. We also show that progressive transfer learning allows us to first learn domain-generic features (for bird classification) which can then be adapted to specific set of bird classes, yielding improvements in accuracy.