Learning Deep Features via Congenerous Cosine Loss for Person Recognition
arXiv:1702.06890
Abstract
Person recognition aims at recognizing the same identity across time and space with complicated scenes and similar appearance. In this paper, we propose a novel method to address this task by training a network to obtain robust and representative features. The intuition is that we directly compare and optimize the cosine distance between two features - enlarging inter-class distinction as well as alleviating inner-class variance. We propose a congenerous cosine loss by minimizing the cosine distance between samples and their cluster centroid in a cooperative way. Such a design reduces the complexity and could be implemented via softmax with normalized inputs. Our method also differs from previous work in person recognition that we do not conduct a second training on the test subset. The identity of a person is determined by measuring the similarity from several body regions in the reference set. Experimental results show that the proposed approach achieves better classification accuracy against previous state-of-the-arts.
Post-rebuttal update. Add some comparison results; correct some technical part; rewrite some sections to make it more readable; code link available
References in corpus (1)
Cited by in corpus (8)
- Support Vector Guided Softmax Loss for Face Recognition
- AdaCos: Adaptively Scaling Cosine Logits for Effectively Learning Deep Face Representations
- Regular Polytope Networks
- Feature Incay for Representation Regularization
- Fix Your Features: Stationary and Maximally Discriminative Embeddings using Regular Polytope (Fixed Classifier) Networks
- Clustering-Oriented Representation Learning with Attractive-Repulsive Loss
- Worse WER, but Better BLEU? Leveraging Word Embedding as Intermediate in Multitask End-to-End Speech Translation
- Exploiting Web Images for Fine-Grained Visual Recognition by Eliminating Noisy Samples and Utilizing Hard Ones