Maximum-Entropy Fine-Grained Classification

Abstract

Fine-Grained Visual Classification (FGVC) is an important computer vision problem that involves small diversity within the different classes, and often requires expert annotators to collect data. Utilizing this notion of small visual diversity, we revisit Maximum-Entropy learning in the context of fine-grained classification, and provide a training routine that maximizes the entropy of the output probability distribution for training convolutional neural networks on FGVC tasks. We provide a theoretical as well as empirical justification of our approach, and achieve state-of-the-art performance across a variety of classification tasks in FGVC, that can potentially be extended to any fine-tuning task. Our method is robust to different hyperparameter values, amount of training data and amount of training label noise and can hence be a valuable tool in many similar problems.

arXiv commentCamera-ready, accepted to NIPS 2018, v2 has minor typo updates and small changes in text

Cited by in corpus (22)

• Transformer with Peak Suppression and Knowledge Guidance for Fine-grained Image Recognition
• Part-guided Relational Transformers for Fine-grained Visual Recognition
• Deep Learning for Fine-Grained Image Analysis: A Survey
• Semi-supervised Domain Adaptation via Minimax Entropy
• Self-Distillation as Instance-Specific Label Smoothing
• Robust Screening of COVID-19 from Chest X-ray via Discriminative Cost-Sensitive Learning
• Regularizing Class-wise Predictions via Self-knowledge Distillation
• Entropic Out-of-Distribution Detection: Seamless Detection of Unknown Examples
• Cross-X Learning for Fine-Grained Visual Categorization
• ProSelfLC: Progressive Self Label Correction for Training Robust Deep Neural Networks
• Attention Convolutional Binary Neural Tree for Fine-Grained Visual Categorization
• Fair Comparison: Quantifying Variance in Resultsfor Fine-grained Visual Categorization
• Fine-grained Recognition: Accounting for Subtle Differences between Similar Classes
• Facing the Hard Problems in FGVC
• Exploring Vision Transformers for Fine-grained Classification
• RAMS-Trans: Recurrent Attention Multi-scale Transformer forFine-grained Image Recognition
• Maximum Entropy Regularization and Chinese Text Recognition
• Fine-Grained Visual Classification via Simultaneously Learning of Multi-regional Multi-grained Features
• Unsupervised Domain Adaptive Object Detection using Forward-Backward Cyclic Adaptation
• Rethinking Uncertainty in Deep Learning: Whether and How it Improves Robustness
• Spatial-Scale Aligned Network for Fine-Grained Recognition
• Salvage Reusable Samples from Noisy Data for Robust Learning