English

Tiny and Efficient Model for the Edge Detection Generalization

Computer Vision and Pattern Recognition 2023-08-15 v1 Machine Learning

Abstract

Most high-level computer vision tasks rely on low-level image operations as their initial processes. Operations such as edge detection, image enhancement, and super-resolution, provide the foundations for higher level image analysis. In this work we address the edge detection considering three main objectives: simplicity, efficiency, and generalization since current state-of-the-art (SOTA) edge detection models are increased in complexity for better accuracy. To achieve this, we present Tiny and Efficient Edge Detector (TEED), a light convolutional neural network with only 58K58K parameters, less than 0.20.2% of the state-of-the-art models. Training on the BIPED dataset takes lessthan30minutesless than 30 minutes, with each epoch requiring lessthan5minutesless than 5 minutes. Our proposed model is easy to train and it quickly converges within very first few epochs, while the predicted edge-maps are crisp and of high quality. Additionally, we propose a new dataset to test the generalization of edge detection, which comprises samples from popular images used in edge detection and image segmentation. The source code is available in https://github.com/xavysp/TEED.

Keywords

Cite

@article{arxiv.2308.06468,
  title  = {Tiny and Efficient Model for the Edge Detection Generalization},
  author = {Xavier Soria and Yachuan Li and Mohammad Rouhani and Angel D. Sappa},
  journal= {arXiv preprint arXiv:2308.06468},
  year   = {2023}
}

Comments

To Appear in ICCV 2023

R2 v1 2026-06-28T11:54:09.786Z