Top-down visual saliency via joint CRF and dictionary learning

By Jimei Yang, Ming-Hsuan Yang. 2012 IEEE Conference on Computer Vision and Pattern Recognition, 2012

458 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/0144b29bde2579e0a1b8ab3a38306c5621a5c30b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Junwei Han
2. Ming-Ming Cheng
3. Huchuan Lu
4. Ling Shao
5. Ming-Hsuan Yang
6. Luc Van Gool
7. Xiaogang Wang
8. G. Wang

Citations in the corpus

• Salient Object Detection: A Discriminative Regional Feature Integration Approach (2013, methodology): “A top-down approach via joint conditional random fields and dictionary learning is introduced [54].”
• Progressively Complementarity-Aware Fusion Network for RGB-D Salient Object Detection (2018, methodology): “Traditional saliency detection models [3, 7-10] are performed merely on RGB images and can be categorized as bottom-up and top-down pipelines.”
• Saliency detection via Cellular Automata (2015, methodology): “Topdown methods [3, 27, 31, 50] are task-driven and require supervised learning with manually labeled ground truth.”
• Saliency detection by multi-context deep learning (2015, methodology): “[58] proposed a top-down saliency model by jointly learning a Conditional Random Field and a dictionary.”
• Salient Object Detection via Structured Matrix Decomposition (2017, unknown): “The latent variable model in [55] estimates saliency by jointly learning a CRF and a specific dictionary.”
• Label Consistent K-SVD: Learning a Discriminative Dictionary for Recognition (2013, methodology): “S PARSE coding has been successfully applied to a variety of problems in computer vision and image analysis, including image denoising [1], image restoration [2], [3], [4], image classification [5], [6], [7], and visual saliency [8].”
• A coarse-to-fine model for airport detection from remote sensing images using target-oriented visual saliency and CRF (2015, methodology): “Specifically, we evaluated a dictionary with 256 atoms learned by the algorithms [24] and a dictionary with 512 atoms learned using the algorithm proposed in [28].”
• Advanced Deep-Learning Techniques for Salient and Category-Specific Object Detection: A Survey (2018, background): “For example, in [50], Yang et al. detected top-down saliency by joint CRF and dictionary learning.”
• Saliency Detection via Absorbing Markov Chain (2013, unknown): “Existing methods are developed with bottom-up visual cues [19, 10, 26, 34] or top-down models [4, 36].”
• Recurrent Attentional Networks for Saliency Detection (2016, methodology): “Top-down methods [48, 46, 26] are based on high-level task-specific prior knowledge.”
• Saliency Detection via Graph-Based Manifold Ranking (2013, unknown): “Bottom-up methods [1, 2, 6, 7, 9– 12, 14, 15, 17, 21, 24, 25, 27, 32, 33, 37] are data-driven and pre-attentive, while top-down methods [23, 36] are taskdriven that entails supervised learning with class labels.”
• BBS-Net: RGB-D Salient Object Detection with a Bifurcated Backbone Strategy Network (2020, methodology): “ree methods, including the adaptive threshold cut (‘ADP’ in the paper), Ostu’s method and the popular algorithm of conditional random fields (CRF). 5.3 Analysis of Post-processing Methods According to [102]–[104], the predicted saliency maps can be further refined by post-processing methods. This may ”
• Object Detection With Deep Learning: A Review (2018, unknown): “object detection, namely bottom-up (BU) [127] and top-down (TD) [128].”