Soria Xavier ; Li Yachuan ; Rouhani Mohammad ; Sappa Angel D.

Abstract

Most high-level computer vision tasks rely on low-level image operations astheir 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 lightconvolutional neural network with only $58K$ parameters, less than $0.2$% ofthe state-of-the-art models. Training on the BIPED dataset takes $less than 30minutes$, with each epoch requiring $less than 5 minutes$. Our proposed modelis easy to train and it quickly converges within very first few epochs, whilethe predicted edge-maps are crisp and of high quality. Additionally, we proposea new dataset to test the generalization of edge detection, which comprisessamples from popular images used in edge detection and image segmentation. Thesource code is available in https://github.com/xavysp/TEED.