One-Shot Adaptation of Supervised Deep Convolutional Models

TL;DR

This paper investigates the extent of dataset bias in deep CNNs trained on large datasets and introduces adaptation methods that improve performance with minimal or no domain-specific labeled data.

Contribution

It demonstrates that large-scale training reduces but does not eliminate dataset bias and proposes novel adaptation techniques effective with limited or no labeled data.

Findings

Deep CNNs trained on large datasets still exhibit dataset bias.

Proposed adaptation methods significantly improve performance on domain adaptation benchmarks.

Adaptation can be effective with as little as one labeled example per category.

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.