Unsupervised Learning of Edges

Abstract

Data-driven approaches for edge detection have proven effective and achieve top results on modern benchmarks. However, all current data-driven edge detectors require manual supervision for training in the form of hand-labeled region segments or object boundaries. Specifically, human annotators mark semantically meaningful edges which are subsequently used for training. Is this form of strong, high-level supervision actually necessary to learn to accurately detect edges? In this work we present a simple yet effective approach for training edge detectors without human supervision. To this end we utilize motion, and more specifically, the only input to our method is noisy semi-dense matches between frames. We begin with only a rudimentary knowledge of edges (in the form of image gradients), and alternate between improving motion estimation and edge detection in turn. Using a large corpus of video data, we show that edge detectors trained using our unsupervised scheme approach the performance of the same methods trained with full supervision (within 3-5%). Finally, we show that when using a deep network for the edge detector, our approach provides a novel pre-training scheme for object detection.

Camera ready version for CVPR 2016

References in corpus (5)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Video (language) modeling: a baseline for generative models of natural videos
• Unsupervised Visual Representation Learning by Context Prediction
• Unsupervised Learning of Visual Representations using Videos
• EpicFlow: Edge-Preserving Interpolation of Correspondences for Optical Flow

Cited by in corpus (5)

• LEGO: Learning Edge with Geometry all at Once by Watching Videos
• Unsupervised Learning for Large-Scale Fiber Detection and Tracking in Microscopic Material Images
• Weakly Supervised Object Boundaries
• Unsupervised Learning of Important Objects from First-Person Videos
• Realtime Hierarchical Clustering based on Boundary and Surface Statistics