Circular Complement Network for RGB-D Salient Object Detection

Bai, Zhen; Li, Zhi; Li, Gongyang; Ye, Linwei; Wang, Yan

doi:10.1016/j.neucom.2021.04.052

Cited by 14 publications

(5 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, PGARNet [8] proposed a progressive guided alternating refinement network for RGB-D salient object detection, which employs recursive and alternating strategies to progressively enhance the accuracy and consistency of saliency prediction. Paper [9] introduced a calibrated RGB-D salient object detection method, which achieves more accurate saliency prediction by jointly learning the calibration mapping between RGB and depth features. CDINet [10] proposed a cross-modal inconsistent interaction network for RGB-D salient object detection.…”

Section: Rgb-d Salient Object Detectionmentioning

confidence: 99%

“…Traditional models based on handcrafted features have limited expressive power, resulting in subpar performance in complex scenes [16], [17], [18], [20]. With the popularity of deep learning, various deep learning-based cross-modal fusion methods have been proposed [8], [9], [10], [11], [12], [13], [19]. For example, literature [36], [38], [44] introduce multiscale fusion methods that fully leverage depth and color information at different scales to improve the performance of salient object detection.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Cross Modal Adaptive Feature Fusion Network for RGB-D Salient Object Detection

Li,

Shi,

Wang

et al. 2024

Preprint

View full text Add to dashboard Cite

RGB-D salient object detection is a fundamental task in computer vision, garnering considerable attention in recent years as many visual tasks rely on salient object detection as a foundational step. However, in the context of RGB-D salient object detection, the effective fusion of RGB modality features with DEPTH modality features to enhance the performance of the salient object detection task remains a significant challenge. In this paper, addressing this issue, we propose an effective network architecture, CMAFF-Net: We design a novel attention mechanism called Normalized Coordinate Attention (NCA) to further extract features from the backbone network, resulting in more effective feature representation. We introduce the CMAFF module to adaptively fuse features extracted at different stages of the backbone network, ensuring the effectiveness of feature fusion at each stage. We propose a hybrid loss function to provide more effective supervision for model learning. Our model exhibits superior performance on the first four publicly available datasets compared to the latest methods. Particularly noteworthy is the approximately 4% improvement in the F-measure evaluation metric on the SSD dataset.

show abstract

Section: Rgb-d Salient Object Detectionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Cross Modal Adaptive Feature Fusion Network for RGB-D Salient Object Detection

Li,

Shi,

Wang

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Despite some improvements, most feature fusion based RGB-D SOD models mentioned above mainly focus on capturing the complementary information within the multi-modality input images, while ignoring the impacts of image qualities on the representation ability of fused features, thus degrading the subsequent saliency detection performance. Recently, some studies have been carried out on the disturbing problem caused by the low-quality images [13], [19]- [21], [54]- [57]. For example, Zhao et al [19] designed a contrast enhancement module with contrast prior information to enhance the quality of depth images, thus boosting the saliency detection performance.…”

Section: B Rgb-d Salient Object Detectionmentioning

confidence: 99%

“…Rather than directly enhancing low-quality depth images, Fan et al [13] proposed a depth depurator unit to reduce the impact of low-quality depth images on the saliency detection performance at the result-level. Bai et al [54] and Li et al [55] employed RGB features to filter distractors in depth features prior to exploiting cross-modal complementarity. Chen et al…”

Section: B Rgb-d Salient Object Detectionmentioning

confidence: 99%

Bi-Directional Progressive Guidance Network for RGB-D Salient Object Detection

Yang

Qin

Luo

et al. 2022

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

“…In the past decades, the visual saliency detection theory [13,14], aiming to highlight the most attractive and distinctive regions in a scene, has been widely used in the field of AD [15,16]. Among them, the context-aware saliency detection [17] is commonly adopted to act as the technique to search for salient objects, due to its powerful performance.…”

Section: Introductionmentioning

confidence: 99%

Hyperspectral Anomaly Detection via Dual Dictionaries Construction Guided by Two-Stage Complementary Decision

et al. 2022

View full text Add to dashboard Cite

Low rank and sparse representation (LRSR) with dual-dictionaries-based methods for detecting anomalies in hyperspectral images (HSIs) are proven to be effective. However, the potential anomaly dictionary is vulnerable to being contaminated by the background pixels in the above methods, and this limits the effect of hyperspectral anomaly detection (HAD). In this paper, a dual dictionaries construction method via two-stage complementary decision (DDC–TSCD) for HAD is proposed. In the first stage, an adaptive inner window–based saliency detection was proposed to yield a coarse binary map, acting as the indicator to select pure background pixels. For the second stage, a background estimation network was designed to generate a fine binary map. Finally, the coarse binary map and fine binary map worked together to construct a pure background dictionary and potential anomaly dictionary in the guidance of the superpixels derived from the first stage. The experiments conducted on public datasets (i.e., HYDICE, Pavia, Los Angeles, San Diego-I, San Diego-II and Texas Coast) demonstrate that DDC–TSCD achieves satisfactory AUC values, which are separately 0.9991, 0.9951, 0.9968, 0.9923, 0.9986 and 0.9969, as compared to four typical methods and three state-of-the-art methods.

show abstract

Circular Complement Network for RGB-D Salient Object Detection

Cited by 14 publications

References 39 publications

Cross Modal Adaptive Feature Fusion Network for RGB-D Salient Object Detection

Cross Modal Adaptive Feature Fusion Network for RGB-D Salient Object Detection

Bi-Directional Progressive Guidance Network for RGB-D Salient Object Detection

Hyperspectral Anomaly Detection via Dual Dictionaries Construction Guided by Two-Stage Complementary Decision

Contact Info

Product

Resources

About