Directional Statistics-based Deep Metric Learning for Image Classification and Retrieval
This work improves image classification and retrieval for computer vision applications by providing a more efficient and effective metric learning approach, though it is incremental as it builds on existing DDML methods.
The paper tackled the problem of deep distance metric learning (DDML) for image tasks by addressing the inaccuracy of Euclidean distance in L2-normalized embedding spaces and the limitations of rigid loss functions based on limited samples, proposing a novel loss function based on the von Mises-Fisher distribution and a new learning algorithm, which achieved state-of-the-art performance on standard datasets with a simpler training procedure and significantly better classification than softmax loss even with fewer layers.
Deep distance metric learning (DDML), which is proposed to learn image similarity metrics in an end-to-end manner based on the convolution neural network, has achieved encouraging results in many computer vision tasks.$L2$-normalization in the embedding space has been used to improve the performance of several DDML methods. However, the commonly used Euclidean distance is no longer an accurate metric for $L2$-normalized embedding space, i.e., a hyper-sphere. Another challenge of current DDML methods is that their loss functions are usually based on rigid data formats, such as the triplet tuple. Thus, an extra process is needed to prepare data in specific formats. In addition, their losses are obtained from a limited number of samples, which leads to a lack of the global view of the embedding space. In this paper, we replace the Euclidean distance with the cosine similarity to better utilize the $L2$-normalization, which is able to attenuate the curse of dimensionality. More specifically, a novel loss function based on the von Mises-Fisher distribution is proposed to learn a compact hyper-spherical embedding space. Moreover, a new efficient learning algorithm is developed to better capture the global structure of the embedding space. Experiments for both classification and retrieval tasks on several standard datasets show that our method achieves state-of-the-art performance with a simpler training procedure. Furthermore, we demonstrate that, even with a small number of convolutional layers, our model can still obtain significantly better classification performance than the widely used softmax loss.