C$^{4}$Net: Contextual Compression and Complementary Combination Network for Salient Object Detection

TL;DR

C$^{4}$Net introduces a novel architecture for salient object detection that emphasizes feature concatenation, joint learning, and specialized modules like CEM and PSM to improve accuracy and edge preservation, outperforming existing methods.

Contribution

The paper proposes a new network architecture with feature concatenation, a Complementary Extraction Module, and Pyramid-Semantic Module, advancing salient object detection performance.

Findings

Outperforms state-of-the-art methods on benchmark datasets

Feature concatenation is more effective than other combination methods

The proposed modules improve edge preservation and detection accuracy

Abstract

Deep learning solutions of the salient object detection problem have achieved great results in recent years. The majority of these models are based on encoders and decoders, with a different multi-feature combination. In this paper, we show that feature concatenation works better than other combination methods like multiplication or addition. Also, joint feature learning gives better results, because of the information sharing during their processing. We designed a Complementary Extraction Module (CEM) to extract necessary features with edge preservation. Our proposed Excessiveness Loss (EL) function helps to reduce false-positive predictions and purifies the edges with other weighted loss functions. Our designed Pyramid-Semantic Module (PSM) with Global guiding flow (G) makes the prediction more accurate by providing high-level complementary information to shallower layers. Experimental results show that the proposed model outperforms the state-of-the-art methods on all benchmark datasets under three evaluation metrics.