Directional Statistics-based Deep Metric Learning for Image Classification and Retrieval

by   Xuefei Zhe, et al.

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.


page 6

page 7

page 8

page 9

page 12


Deep Metric Learning via Facility Location

Learning the representation and the similarity metric in an end-to-end f...

Human Motion Analysis with Deep Metric Learning

Effectively measuring the similarity between two human motions is necess...

Dissecting Deep Metric Learning Losses for Image-Text Retrieval

Visual-Semantic Embedding (VSE) is a prevalent approach in image-text re...

Learning Deep Optimal Embeddings with Sinkhorn Divergences

Deep Metric Learning algorithms aim to learn an efficient embedding spac...

Domain-aware Triplet loss in Domain Generalization

Despite much progress being made in the field of object recognition with...

von Mises-Fisher Loss: An Exploration of Embedding Geometries for Supervised Learning

Recent work has argued that classification losses utilizing softmax cros...

Beyond Correlation: A Path-Invariant Measure for Seismogram Similarity

Similarity search is a popular technique for seismic signal processing, ...

Please sign up or login with your details

Forgot password? Click here to reset