Practical Edge Detection via Robust Collaborative Learning

TL;DR

This paper introduces PEdger, a robust and efficient edge detection model that leverages collaborative learning from different training moments and architectures to improve accuracy and robustness without relying on pre-trained backbones.

Contribution

The paper proposes a novel collaborative learning framework for edge detection that reduces reliance on pre-trained models and mitigates noisy label effects, enhancing efficiency and robustness.

Findings

Outperforms existing methods in accuracy, speed, and model size.

Effective in handling noisy annotations without extra pre-training.

Validated on BSDS500 and NYUD datasets.

Abstract

Edge detection, as a core component in a wide range of visionoriented tasks, is to identify object boundaries and prominent edges in natural images. An edge detector is desired to be both efficient and accurate for practical use. To achieve the goal, two key issues should be concerned: 1) How to liberate deep edge models from inefficient pre-trained backbones that are leveraged by most existing deep learning methods, for saving the computational cost and cutting the model size; and 2) How to mitigate the negative influence from noisy or even wrong labels in training data, which widely exist in edge detection due to the subjectivity and ambiguity of annotators, for the robustness and accuracy. In this paper, we attempt to simultaneously address the above problems via developing a collaborative learning based model, termed PEdger. The principle behind our PEdger is that, the information learned from different training moments and heterogeneous (recurrent and non recurrent in this work) architectures, can be assembled to explore robust knowledge against noisy annotations, even without the help of pre-training on extra data. Extensive ablation studies together with quantitative and qualitative experimental comparisons on the BSDS500 and NYUD datasets are conducted to verify the effectiveness of our design, and demonstrate its superiority over other competitors in terms of accuracy, speed, and model size. Codes can be found at https://github.co/ForawardStar/PEdger.