LFNet: Light Field Fusion Network for Salient Object Detection

Zhang, Miao; Ji, Wei; Piao, Yongri; Li, Jingjing; Zhang, Yu; Xu, Shuang; Lu, Huchuan

doi:10.1109/tip.2020.2990341

Cited by 101 publications

(49 citation statements)

References 67 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The MA method [145] employs a two-stage saliency refinement strategy to produce the final prediction map, so that adjacent superpixels obtain similar saliency values. LFNet [141] presents an effective refinement module to reduce the homogeneity between different modalities as well to refine their dissimilarities.…”

Section: Refinement-based Modelsmentioning

confidence: 99%

See 1 more Smart Citation

RGB-D salient object detection: A survey

et al. 2021

View full text Add to dashboard Cite

Salient object detection, which simulates human visual perception in locating the most significant object(s) in a scene, has been widely applied to various computer vision tasks. Now, the advent of depth sensors means that depth maps can easily be captured; this additional spatial information can boost the performance of salient object detection. Although various RGB-D based salient object detection models with promising performance have been proposed over the past several years, an in-depth understanding of these models and the challenges in this field remains lacking. In this paper, we provide a comprehensive survey of RGB-D based salient object detection models from various perspectives, and review related benchmark datasets in detail. Further, as light fields can also provide depth maps, we review salient object detection models and popular benchmark datasets from this domain too. Moreover, to investigate the ability of existing models to detect salient objects, we have carried out a comprehensive attribute-based evaluation of several representative RGB-D based salient object detection models. Finally, we discuss several challenges and open directions of RGB-D based salient object detection for future research. All collected models, benchmark datasets, datasets constructed for attribute-based evaluation, and related code are publicly available at https://github.com/taozh2017/RGBD-SODsurvey.

show abstract

Section: Refinement-based Modelsmentioning

confidence: 99%

“…Salient object detection methods can be grouped into three categories according to the input data type: RGB, RGB-D, or light field [141]. We have already reviewed RGB-D based salient object detection models, in which depth maps provide geometric information to improve salient object detection performance to some extent.…”

Section: Introductionmentioning

confidence: 99%

RGB-D salient object detection: A survey

et al. 2021

View full text Add to dashboard Cite

show abstract

“…In recent years, convolutional neural networks (CNNs) based SOD methods [12,18,20,21,30,36,42,46,47,[54][55][56][57]62] have obtained vast success and pushed the performance of SOD to a new level. In CNN-based SOD models, features at different levels can represent different characteristics of salient objects.…”

Section: Introductionmentioning

confidence: 99%

Auto-MSFNet: Search Multi-scale Fusion Network for Salient Object Detection

Zhang

Liu

Piao

et al. 2021

Proceedings of the 29th ACM International Conference on Multimedia

Self Cite

View full text Add to dashboard Cite

Multi-scale features fusion plays a critical role in salient object detection. Most of existing methods have achieved remarkable performance by exploiting various multi-scale features fusion strategies. However, an elegant fusion framework requires expert knowledge and experience, heavily relying on laborious trial and error. In this paper, we propose a multi-scale features fusion framework based on Neural Architecture Search (NAS), named Auto-MSFNet. First, we design a novel search cell, named FusionCell to automatically decide multi-scale features aggregation. Rather than searching one repeatable cell stacked, we allow different FusionCells to flexibly integrate multi-level features. Simultaneously, considering features generated from CNNs are naturally spatial and channel-wise, we propose a new search space for efficiently focusing on the most relevant information. The search space mitigates incomplete object structures or over-predicted foreground regions caused by progressive fusion. Second, we propose a progressive polishing loss to further obtain exquisite boundaries by penalizing misalignment of salient object boundaries. Extensive experiments on five benchmark datasets demonstrate the effectiveness of the proposed method and achieve state-of-the-art performance on four evaluation metrics. The code and results of our method are available at https://github.com/OIPLab-DUT/Auto-MSFNet. CCS CONCEPTS• Computing methodologies → Interest point and salient region detections.

show abstract

“…In contrast, numerous salient object detection (SOD) methods have been developed in computer vision (CV) [22], [23], [24], [25], [26], [27], [28]. SOD is a task based on a visual attention mechanism, which aims to detect objects more attractive than the surrounding areas in a scene or an image [29], [30].…”

Section: Introductionmentioning

confidence: 99%

ESPFNet: An Edge-Aware Spatial Pyramid Fusion Network for Salient Shadow Detection in Aerial Remote Sensing Images

Luo

Zhu³

et al. 2021

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Shadows can hinder image interpretation in aerial remote sensing images. The existing shadow detection methods focus on all shadow regions and detect the shadow regions directly, but they ignore the fact that salient shadows have a more significant effect. In this work, a novel edge-aware spatial pyramid fusion network (ESPFNet) under a multi-task learning framework is proposed for salient shadow detection in aerial remote sensing images. ESPFNet has three components: 1) a parallel spatial pyramid (PSP) structure; 2) an edge detection module (EDM); and 3) an edge-aware multi-branch integration (EMI). The PSP structure is constructed to extract multi-scale features from the input image and fuse them gradually. The EDM then integrates the shallow features and deep features to detect the shadow edges. Finally, the EMI incorporates the edge features with multi-branch features, and then concatenates them with the shallow features to generate the salient shadow detection result. The experimental analyses confirm the effectiveness of the ESPFNet method in both the qualitative and quantitative performance, compared to the existing methods, with the F-score reaching 92.04% in the salient shadow test set.

show abstract

LFNet: Light Field Fusion Network for Salient Object Detection

Cited by 101 publications

References 67 publications

RGB-D salient object detection: A survey

RGB-D salient object detection: A survey

Auto-MSFNet: Search Multi-scale Fusion Network for Salient Object Detection

ESPFNet: An Edge-Aware Spatial Pyramid Fusion Network for Salient Shadow Detection in Aerial Remote Sensing Images

Contact Info

Product

Resources

About