Compact Twice Fusion Network for Edge Detection

TL;DR

This paper introduces a compact, efficient edge detection network that effectively fuses multi-scale features using lightweight modules and a novel loss function, achieving competitive accuracy with fewer parameters.

Contribution

The paper proposes a novel Compact Twice Fusion Network with lightweight modules and a Dynamic Focal Loss for improved edge detection efficiency and accuracy.

Findings

Achieves competitive accuracy on BSDS500, NYUDv2, BIPEDv2 datasets.

Uses only 0.1M additional parameters, reducing computation to 60%.

Outperforms state-of-the-art methods in efficiency and accuracy.

Abstract

The significance of multi-scale features has been gradually recognized by the edge detection community. However, the fusion of multi-scale features increases the complexity of the model, which is not friendly to practical application. In this work, we propose a Compact Twice Fusion Network (CTFN) to fully integrate multi-scale features while maintaining the compactness of the model. CTFN includes two lightweight multi-scale feature fusion modules: a Semantic Enhancement Module (SEM) that can utilize the semantic information contained in coarse-scale features to guide the learning of fine-scale features, and a Pseudo Pixel-level Weighting (PPW) module that aggregate the complementary merits of multi-scale features by assigning weights to all features. Notwithstanding all this, the interference of texture noise makes the correct classification of some pixels still a challenge. For these hard samples, we propose a novel loss function, coined Dynamic Focal Loss, which reshapes the standard cross-entropy loss and dynamically adjusts the weights to correct the distribution of hard samples. We evaluate our method on three datasets, i.e., BSDS500, NYUDv2, and BIPEDv2. Compared with state-of-the-art methods, CTFN achieves competitive accuracy with less parameters and computational cost. Apart from the backbone, CTFN requires only 0.1M additional parameters, which reduces its computation cost to just 60% of other state-of-the-art methods. The codes are available at https://github.com/Li-yachuan/CTFN-pytorch-master.