Rethinking Detecting Salient and Camouflaged Objects in Unconstrained Scenes

TL;DR

This paper introduces USC12K, a large-scale dataset and USCNet, a model with explicit attribute relationship modeling, to improve the detection of salient and camouflaged objects in unconstrained scenes, addressing limitations of existing datasets and models.

Contribution

The paper presents a new dataset USC12K and a novel model USCNet that explicitly models relationships between salient and camouflaged objects for better detection.

Findings

USCNet achieves state-of-the-art performance across all scenes.

The dataset USC12K covers all logical existence scenarios of objects.

The proposed evaluation metric CSCS effectively assesses model discrimination.

Abstract

While the human visual system employs distinct mechanisms to perceive salient and camouflaged objects, existing models struggle to disentangle these tasks. Specifically, salient object detection (SOD) models frequently misclassify camouflaged objects as salient, while camouflaged object detection (COD) models conversely misinterpret salient objects as camouflaged. We hypothesize that this can be attributed to two factors: (i) the specific annotation paradigm of current SOD and COD datasets, and (ii) the lack of explicit attribute relationship modeling in current models. Prevalent SOD/COD datasets enforce a mutual exclusivity constraint, assuming scenes contain either salient or camouflaged objects, which poorly aligns with the real world. Furthermore, current SOD/COD methods are primarily designed for these highly constrained datasets and lack explicit modeling of the relationship between salient and camouflaged objects. In this paper, to promote the development of unconstrained salient and camouflaged object detection, we construct a large-scale dataset, USC12K, which features comprehensive labels and four different scenes that cover all possible logical existence scenarios of both salient and camouflaged objects. To explicitly model the relationship between salient and camouflaged objects, we propose a model called USCNet, which introduces two distinct prompt query mechanisms for modeling inter-sample and intra-sample attribute relationships. Additionally, to assess the model's ability to distinguish between salient and camouflaged objects, we design an evaluation metric called CSCS. The proposed method achieves state-of-the-art performance across all scenes in various metrics. The code and dataset will be available at https://github.com/ssecv/USCNet.