Salience Guided Depth Calibration for Perceptually Optimized Compressive Light Field 3D Display

Wang, Shizheng; Liao, Wenjuan; Surman, Phil; Tu, Zhigang; Zheng, Yuanjin; Yuan, Junsong

doi:10.1109/cvpr.2018.00217

Cited by 16 publications

(12 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, proposed a unified architecture based on weighted sparse coding. Later methods [Zhang et al, 2015;Sheng et al, 2016;Wang et al, 2017a;Zhang et al, 2017a;Wang et al, 2018a;Wang et al, 2018b] explored and further combined multiple visual cues (e.g., depth, color contrast, light field flows and boundary prior) to detect saliency. Most recent methods shifted more attention to depth information and employed cellular automata for the saliency detection in light field.…”

Section: Light Field Sodmentioning

confidence: 99%

“…Generally, there are three main categories of attention mechanisms, including: channel attention , spatial attention [Woo et al, 2018], and self-attention [Wang et al, 2018c], which is a concept borrowed from the field of natural language processing. These attention mechanisms can be easily embedded to different CNN-based architectures.…”

Section: Attention Mechanismmentioning

confidence: 99%

“…Considering the variations of spatial information among each of the inputting image scenes, a self-gate mechanism [Wang et al, 2018c] is further employed to automatically learn the co-attention confidences…”

Section: Synergistic Attention Modulementioning

confidence: 99%

See 2 more Smart Citations

Learning Synergistic Attention for Light Field Salient Object Detection

Zhang¹,

Chen²,

Qian³

et al. 2021

Preprint

View full text Add to dashboard Cite

We propose a novel Synergistic Attention Network (SA-Net) to address the light field salient object detection by establishing a synergistic effect between multi-modal features with advanced attention mechanisms. Our SA-Net exploits the rich information of focal stacks via 3D convolutional neural networks, decodes the high-level features of multi-modal light field data with two cascaded synergistic attention modules, and predicts the saliency map using an effective feature fusion module in a progressive manner. Extensive experiments on three widely-used benchmark datasets show that our SA-Net outperforms 28 state-of-the-art models, sufficiently demonstrating its effectiveness and superiority. Our code will be made publicly available at http://hidden.for.anonymity.

show abstract

Section: Light Field Sodmentioning

confidence: 99%

Section: Attention Mechanismmentioning

confidence: 99%

See 1 more Smart Citation

Learning Synergistic Attention for Light Field Salient Object Detection

Zhang¹,

Chen²,

Qian³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Traditional/Deep Models. The classic models for light field SOD often use superpixel-level hand-crafted features [95], [129]- [135], [137], [143]. Early work [129], [135] showed that the unique refocusing capability of light fields can provide useful focusness, depth, and objectness cues, and further proposed several SOD models using light field data.…”

Section: Saliency Detection On Light Field a Light Field Sod Modelsmentioning

confidence: 99%

RGB-D Salient Object Detection: A Survey

Zhou,

Fan,

Cheng

et al. 2020

Preprint

View full text Add to dashboard Cite

“…In this paper, we just make use of the camera depth region, [LambdaMin, LambdaMax], to guide the proposed optimization. All the depth parameters are extracted from the public dataset [13] and shared in the supplementary material of our related paper [12].…”

Section: Light Field Capture and Analysismentioning

confidence: 99%

P‐200: Late‐News Poster: Depth Guided and Blur Aware Optimization for Multi‐Layer Light Field 3D Display

Feng

Surman

et al. 2019

Symp Digest of Tech Papers

Self Cite

View full text Add to dashboard Cite

Multi-layer light field 3D displays are a type of computational 3D display which are gaining increasing interest for their holographiclike effect and natural compatibility with 2D displays. However, two major shortcomings; depth limitation and multi-layer blur are usually ignored in the traditional light field modeling and reconstruction based on liquid crystal displays (LCDs). Considering these two disadvantages, this paper proposes a depth guided and blur-dependent optimization for multi-layer light field displays. Different from the previous cascaded light field displays, which only use the fixed initialization plane as the depth center of the display content, the proposed method automatically chooses the best depth initialization from three different classical configurations by introducing the capture depth of camera imaging and a weighted blur factor of multi-layer display into the optimization. Experiments demonstrate that the proposed method provides a promising advantage in visual perception of compressive light field displays with both software simulation and prototype demonstration.

show abstract

Salience Guided Depth Calibration for Perceptually Optimized Compressive Light Field 3D Display

Cited by 16 publications

References 43 publications

Learning Synergistic Attention for Light Field Salient Object Detection

Learning Synergistic Attention for Light Field Salient Object Detection

RGB-D Salient Object Detection: A Survey

P‐200: Late‐News Poster: Depth Guided and Blur Aware Optimization for Multi‐Layer Light Field 3D Display

Contact Info

Product

Resources

About