CVMay 19, 2015

Unsupervised Visual Representation Learning by Context Prediction

arXiv:1505.05192v32969 citations
Originality Highly original
AI Analysis

It addresses the problem of learning visual representations without labeled data for computer vision tasks, offering a novel unsupervised approach that is not incremental.

The paper tackles unsupervised visual representation learning by using spatial context prediction as a supervisory signal, training a convolutional neural net to predict relative patch positions from unlabeled images, which enables unsupervised object discovery and achieves state-of-the-art performance on Pascal VOC 2011 when used in the R-CNN framework.

This work explores the use of spatial context as a source of free and plentiful supervisory signal for training a rich visual representation. Given only a large, unlabeled image collection, we extract random pairs of patches from each image and train a convolutional neural net to predict the position of the second patch relative to the first. We argue that doing well on this task requires the model to learn to recognize objects and their parts. We demonstrate that the feature representation learned using this within-image context indeed captures visual similarity across images. For example, this representation allows us to perform unsupervised visual discovery of objects like cats, people, and even birds from the Pascal VOC 2011 detection dataset. Furthermore, we show that the learned ConvNet can be used in the R-CNN framework and provides a significant boost over a randomly-initialized ConvNet, resulting in state-of-the-art performance among algorithms which use only Pascal-provided training set annotations.

Code Implementations3 repos
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes