Camouflaged Object Detection with Feature Grafting and Distractor Aware

TL;DR

This paper introduces FDNet, a novel neural network combining CNN and Transformer encoders with feature grafting and distractor modeling to improve camouflaged object detection, supported by a new large dataset and extensive experiments.

Contribution

The paper proposes a new FDNet architecture with feature grafting and distractor awareness for improved camouflaged object detection, along with the largest dataset ACOD2K.

Findings

FDNet outperforms state-of-the-art methods on multiple benchmarks.

The cross-attention feature grafting enhances feature integration.

The distractor aware module refines camouflage detection accuracy.

Abstract

The task of Camouflaged Object Detection (COD) aims to accurately segment camouflaged objects that integrated into the environment, which is more challenging than ordinary detection as the texture between the target and background is visually indistinguishable. In this paper, we proposed a novel Feature Grafting and Distractor Aware network (FDNet) to handle the COD task. Specifically, we use CNN and Transformer to encode multi-scale images in parallel. In order to better explore the advantages of the two encoders, we design a cross-attention-based Feature Grafting Module to graft features extracted from Transformer branch into CNN branch, after which the features are aggregated in the Feature Fusion Module. A Distractor Aware Module is designed to explicitly model the two possible distractors in the COD task to refine the coarse camouflage map. We also proposed the largest artificial camouflaged object dataset which contains 2000 images with annotations, named ACOD2K. We conducted extensive experiments on four widely used benchmark datasets and the ACOD2K dataset. The results show that our method significantly outperforms other state-of-the-art methods. The code and the ACOD2K will be available at https://github.com/syxvision/FDNet.