Local Feature Descriptor Learning with Adaptive Siamese Network

Fig.1. The pipeline of our proposed method. (a) Data augmentation. (b) Generate labeled pairs for training. (c) Learn feature descriptors using Siamese architecture.

Fig. 2. (left) Siamese Architecture (right) The activation frequency of a fully-connected layer given 100k patches.

Although the recent progress in the deep neural network has led to the development of learnable local feature descriptors, there is no explicit answer for estimation of the necessary size of a neural network. Specifically, the local feature is represented in a low dimensional space, so the neural network should have more compact structure. The small networks required for local feature descriptor learning may be sensitive to initial conditions and learning parameters and more likely to become trapped in local minima. In order to address the above problem, we introduce an adaptive pruning Siamese Architecture based on neuron activation to learn local feature descriptors, making the network more computationally efficient with an improved recognition rate over more complex networks. Our experiments demonstrate that our learned local feature descriptors outperform the state-of-art methods in patch matching.

Publication:

C. Huang, Q. Liu, Y. -Y. Chen and K. -T. Cheng, "Local Feature Descriptor Learning with Adaptive Siamese Network," in Local Features: State of the art, open problems and performance evaluation (In conjunction with ECCV), 2016.