One-Shot Adaptation of Supervised Deep Convolutional Models

Abstract

Dataset bias remains a significant barrier towards solving real world computer vision tasks. Though deep convolutional networks have proven to be a competitive approach for image classification, a question remains: have these models have solved the dataset bias problem? In general, training or fine-tuning a state-of-the-art deep model on a new domain requires a significant amount of data, which for many applications is simply not available. Transfer of models directly to new domains without adaptation has historically led to poor recognition performance. In this paper, we pose the following question: is a single image dataset, much larger than previously explored for adaptation, comprehensive enough to learn general deep models that may be effectively applied to new image domains? In other words, are deep CNNs trained on large amounts of labeled data as susceptible to dataset bias as previous methods have been shown to be? We show that a generic supervised deep CNN model trained on a large dataset reduces, but does not remove, dataset bias. Furthermore, we propose several methods for adaptation with deep models that are able to operate with little (one example per category) or no labeled domain specific data. Our experiments show that adaptation of deep models on benchmark visual domain adaptation datasets can provide a significant performance boost.

References in corpus (3)

• DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition
• Frustratingly Easy Domain Adaptation
• Visualizing and Understanding Convolutional Networks

Cited by in corpus (18)

• Unsupervised Domain Adaptation by Backpropagation
• Deep Visual Domain Adaptation: A Survey
• Generalizing from a Few Examples: A Survey on Few-Shot Learning
• Hierarchical Multiscale Recurrent Neural Networks
• Convolutional Neural Network-based Place Recognition
• Deep Reconstruction-Classification Networks for Unsupervised Domain Adaptation
• Fast Rates by Transferring from Auxiliary Hypotheses
• Are we done with object recognition? The iCub robot's perspective
• When Unsupervised Domain Adaptation Meets Tensor Representations
• Deep Learning for Single-View Instance Recognition
• Online and Offline Handwritten Chinese Character Recognition: A Comprehensive Study and New Benchmark
• Joint cross-domain classification and subspace learning for unsupervised adaptation
• A Brain-like Cognitive Process with Shared Methods
• Self-Supervised Dynamic Networks for Covariate Shift Robustness
• A New Oscillating-Error Technique for Classifiers
• Evaluation of Deep Neural Network Domain Adaptation Techniques for Image Recognition
• Localising In Complex Scenes Using Balanced Adversarial Adaptation
• Cross-domain Image Retrieval with a Dual Attribute-aware Ranking Network