Prototype Memory for Large-scale Face Representation Learning
Explore this paper's citation graph
Summary
A novel face representation learning model called Prototype Memory, which alleviates "prototype obsolescence" and allows training on a dataset of any size and can be used with various loss functions, hard example mining algorithms and encoder architectures.
- Type
- article
- Published
- 2021-05-05
- Cited by
- 5
- References
- 147
- Access
- Open access
- OpenAlex
- https://openalex.org/W3157066609
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:233739969
Keywords
Computer science, Scale (ratio), Artificial intelligence, Face (sociological concept), Representation (politics)
References
- Labeled Faces in the Wild: A Database forStudying Face Recognition in Unconstrained Environments
- Distilling the Knowledge in a Neural Network
- FaceNet: A unified embedding for face recognition and clustering
- DeepFace: Closing the Gap to Human-Level Performance in Face Verification
- Learning a similarity metric discriminatively, with application to face verification
- The MegaFace Benchmark: 1 Million Faces for Recognition at Scale
- Deep Residual Learning for Image Recognition
- Deep Face Recognition
- Joint Face Detection and Alignment Using Multitask Cascaded Convolutional Networks
- Frontal to profile face verification in the wild
- Deep Convolutional Neural Network with Independent Softmax for Large Scale Face Recognition
- UMDFaces: An annotated face dataset for training deep networks
- Prototypical Networks for Few-shot Learning
- NormFace: L2 Hypersphere Embedding for Face Verification
- Multi-task Deep Neural Network for Joint Face Recognition and Facial Attribute Prediction
- AgeDB: The First Manually Collected, In-the-Wild Age Database
- TAPAS: Two-pass Approximate Adaptive Sampling for Softmax
- Marginal Loss for Deep Face Recognition
- Self-organized Hierarchical Softmax
- Rethinking Feature Discrimination and Polymerization for Large-scale Recognition
Cited by
- STC speaker recognition systems for the NIST SRE 2021
- Quality-Aware Prototype Memory for Face Representation Learning
- Efficient Extreme Large-Scale Speaker Verification: Dynamic Active Sub Fully-Connected Layers for Faster Training and Memory Optimization
- Second FRCSyn-onGoing: Winning Solutions and Post-Challenge Analysis to Improve Face Recognition with Synthetic Data
- MCP: Momentum-Based Prototype Regularization for Compact and Separable Visual Representations
Related papers
- Elections and Representation in Local Government: A Victorian Case Study
- Prelude to a Theory of Musical Representation
- Women’s Representation in Argentine National and Subnational Governments
- Political Argument about Representation: The Case of the Maori Seats
- What Do We See In Film
- Female Strategies of Self-representation in Variants, Versions, and Revisions of “Bluebeard”
- Representation and Reference
- The Advantages and Reform of Part-time Representation of China's People's Congress
- The Representation of False Prophets in Polemical Discourses
- New perspectives on power and political representation from ancient history to the present day: repertoires of representation