Improved Deep Metric Learning with Multi-class N-pair Loss Objective
Explore this paper's citation graph
Summary
This paper proposes a new metric learning objective called multi-class N-pair loss, which generalizes triplet loss by allowing joint comparison among more than one negative examples and reduces the computational burden of evaluating deep embedding vectors via an efficient batch construction strategy using only N pairs of examples.
- Type
- article
- Published
- 2016-12-05
- Cited by
- 2,505
- References
- 32
- OpenAlex
- https://openalex.org/W2555897561
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:911406
Keywords
Deep learning, Metric (unit), Computer science, Embedding, Benchmark (surveying)
References
- Learning Face Representation from Scratch
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Targeting Ultimate Accuracy: Face Recognition via Deep Embedding
- The Caltech-UCSD Birds-200-2011 Dataset
- Hyper-class augmented and regularized deep learning for fine-grained image classification
- Learning Fine-Grained Image Similarity with Deep Ranking
- Similarity Metric Learning for a Variable-Kernel Classifier
- Unconstrained Face Recognition: Identifying a Person of Interest From a Media Collection
- FaceNet: A unified embedding for face recognition and clustering
- Going deeper with convolutions
- Distance Metric Learning for Large Margin Nearest Neighbor Classification
- Hamming Distance Metric Learning
- Distance Metric Learning with Application to Clustering with Side-Information
- Product Quantization for Nearest Neighbor Search
- 3D Object Representations for Fine-Grained Categorization
- Dimensionality Reduction by Learning an Invariant Mapping
- Neighbourhood Components Analysis
- DeepFace: Closing the Gap to Human-Level Performance in Face Verification
- Caffe: Convolutional Architecture for Fast Feature Embedding
- Learning a similarity metric discriminatively, with application to face verification
Cited by
- What Makes Objects Similar: A Unified Multi-Metric Learning Approach
- Hard-Aware Deeply Cascaded Embedding
- Learnable Structured Clustering Framework for Deep Metric Learning
- Multi-Task Convolutional Neural Network for Pose-Invariant Face Recognition
- Reconstruction for Feature Disentanglement in Pose-invariant Face Recognition
- No Fuss Distance Metric Learning Using Proxies
- Semantic Instance Segmentation via Deep Metric Learning
- Kernelized evolutionary distance metric learning for semi-supervised clustering
- Smart Mining for Deep Metric Learning
- NormFace: L2 Hypersphere Embedding for Face Verification
- Nearest Neighbour Radial Basis Function Solvers for Deep Neural Networks
- Sampling Matters in Deep Embedding Learning
- Analyse fine 2D/3D de véhicules par réseaux de neurones profonds. (2D/3D fine-grained analysis of vehicles using deep neural networks)
- Adaptive Deep Metric Learning for Identity-Aware Facial Expression Recognition
- Deep Spectral Clustering Learning
- Unsupervised Domain Adaptation for Face Recognition in Unlabeled Videos
- Deep Metric Learning with Angular Loss
- Cross-Domain Shoe Retrieval With a Semantic Hierarchy of Attribute Classification Network
- Representation Learning for Visual-Relational Knowledge Graphs
- Metric-based Generative Adversarial Network