Discriminative Unsupervised Feature Learning with Exemplar Convolutional Neural Networks

Abstract

Deep convolutional networks have proven to be very successful in learning task specific features that allow for unprecedented performance on various computer vision tasks. Training of such networks follows mostly the supervised learning paradigm, where sufficiently many input-output pairs are required for training. Acquisition of large training sets is one of the key challenges, when approaching a new task. In this paper, we aim for generic feature learning and present an approach for training a convolutional network using only unlabeled data. To this end, we train the network to discriminate between a set of surrogate classes. Each surrogate class is formed by applying a variety of transformations to a randomly sampled 'seed' image patch. In contrast to supervised network training, the resulting feature representation is not class specific. It rather provides robustness to the transformations that have been applied during training. This generic feature representation allows for classification results that outperform the state of the art for unsupervised learning on several popular datasets (STL-10, CIFAR-10, Caltech-101, Caltech-256). While such generic features cannot compete with class specific features from supervised training on a classification task, we show that they are advantageous on geometric matching problems, where they also outperform the SIFT descriptor.

PAMI submission. Includes matching experiments as in arXiv:1405.5769v1. Also includes new network architectures, experiments on Caltech-256, experiment on combining Exemplar-CNN with clustering

References in corpus (7)

• Improving neural networks by preventing co-adaptation of feature detectors
• Natural Language Processing (almost) from Scratch
• Caffe: Convolutional Architecture for Fast Feature Embedding
• DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition
• Deeply-Supervised Nets
• Dropout Training as Adaptive Regularization
• Descriptor Matching with Convolutional Neural Networks: a Comparison to SIFT

Cited by in corpus (12)

• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Learning to Compose Domain-Specific Transformations for Data Augmentation
• Loss-Sensitive Generative Adversarial Networks on Lipschitz Densities
• Patch2Self: Denoising Diffusion MRI with Self-Supervised Learning
• Ultrasound Image Representation Learning by Modeling Sonographer Visual Attention
• Unsupervised Feature Learning by Cross-Level Instance-Group Discrimination
• DisCo: Remedy Self-supervised Learning on Lightweight Models with Distilled Contrastive Learning
• Mix-and-Match Tuning for Self-Supervised Semantic Segmentation
• Improving Context-Based Meta-Reinforcement Learning with Self-Supervised Trajectory Contrastive Learning
• Unsupervised Object Detection with LiDAR Clues
• FROST: Faster and more Robust One-shot Semi-supervised Training
• Don't Wait, Just Weight: Improving Unsupervised Representations by Learning Goal-Driven Instance Weights