Unsupervised Embedding Learning via Invariant and Spreading Instance Feature

Abstract

This paper studies the unsupervised embedding learning problem, which requires an effective similarity measurement between samples in low-dimensional embedding space. Motivated by the positive concentrated and negative separated properties observed from category-wise supervised learning, we propose to utilize the instance-wise supervision to approximate these properties, which aims at learning data augmentation invariant and instance spread-out features. To achieve this goal, we propose a novel instance based softmax embedding method, which directly optimizes the `real' instance features on top of the softmax function. It achieves significantly faster learning speed and higher accuracy than all existing methods. The proposed method performs well for both seen and unseen testing categories with cosine similarity. It also achieves competitive performance even without pre-trained network over samples from fine-grained categories.

CVPR 2019

References in corpus (8)

• Distilling the Knowledge in a Neural Network
• In Defense of the Triplet Loss for Person Re-Identification
• Classification with Noisy Labels by Importance Reweighting
• Adversarial Feature Learning
• Adversarially Learned Inference
• AlignedReID: Surpassing Human-Level Performance in Person Re-Identification
• Stacked What-Where Auto-encoders
• Significance of Softmax-based Features in Comparison to Distance Metric Learning-based Features

Cited by in corpus (19)

• Momentum Contrast for Unsupervised Visual Representation Learning
• Exploring Simple Siamese Representation Learning
• Deep Graph Contrastive Representation Learning
• Robust Pre-Training by Adversarial Contrastive Learning
• i-Mix: A Domain-Agnostic Strategy for Contrastive Representation Learning
• Contrastive Learning with Adversarial Examples
• Uncertainty Quantification and Deep Ensembles
• Delving Deeper into Anti-aliasing in ConvNets
• Unsupervised Feature Learning by Cross-Level Instance-Group Discrimination
• Exploring Instance Relations for Unsupervised Feature Embedding
• Unsupervised Deep Metric Learning via Auxiliary Rotation Loss
• Joint Contrastive Learning with Infinite Possibilities
• CO2: Consistent Contrast for Unsupervised Visual Representation Learning
• Deep Metric Learning Meets Deep Clustering: An Novel Unsupervised Approach for Feature Embedding
• CSNNs: Unsupervised, Backpropagation-free Convolutional Neural Networks for Representation Learning
• Acceleration of Large Margin Metric Learning for Nearest Neighbor Classification Using Triplet Mining and Stratified Sampling
• Unsupervised Place Recognition with Deep Embedding Learning over Radar Videos
• Reliable Label Bootstrapping for Semi-Supervised Learning
• Exploit Clues from Views: Self-Supervised and Regularized Learning for Multiview Object Recognition