Segmentation guided local proposal fusion for co-saliency detection

Tsai, Chung-Chi; Qian, Xiaoning; Lin, Yen‐Yu

doi:10.1109/icme.2017.8019413

Cited by 12 publications

(10 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…See the second and third columns of cows on MSRC v2 dataset. The backgrounds contain grass, which affect the detection of some compared methods such as CSHS [17], CoDW [16], SACS [27] and SGCS [30]. Comparatively, our method can better separate common objects and backgrounds with clearer boundaries.…”

Section: E Comparison With Baselinesmentioning

confidence: 99%

“…We compare our method with representative approaches including CBCS [13], CBCS-s [13], CoDW [16], CSCO [31], CSHS [17], DIM [21], ESMG [18], GwD [35], IRSD [28], IRSD-s [28], SP-MIL [34], UMLBF [50], SACS [27], SACSs [27], SGCS [30], UCSG [83] and Gw-FCN [36]. Among them, CBCS [13], CoDW [16], CSCO [31], CSHS [17], ESMG [18], IRSD [28], SP-MIL [34], SACS [27], SGCS [30] and UCSG [83] are unsupervised co-saliency methods; DIM [21], GwD [35], UMLBF [50] and Gw-FCN [36] are supervised co-saliency methods; to investigate the influence of interaction information, CBCS-s [13], IRSD-s [28], SACSs [27], which are aimed at single image saliency detection, are also used as baselines in our work. When available, we use the publicly released source code with default parameters provided by the authors to reproduce the experiments on our test sets.…”

Section: Compared Baselinesmentioning

confidence: 99%

“…To sum up, compared with the unsupervised co-saliency detection methods [13], [16]- [18], [27], [28], [30], [31], [34], our proposed method can find more complete co-salient objects; compared with the supervised co-saliency detection methods [21], [50], it can retain more clearer backgrounds; compared with the methods for single image saliency detection, it can make the co-salient objects more highlighted and the backgrounds more suppressed. These results and analyses well speak for the effectiveness of our method.…”

Section: E Comparison With Baselinesmentioning

confidence: 99%

See 2 more Smart Citations

Co-Saliency Detection With Co-Attention Fully Convolutional Network

Gao

Zhao²,

Liu

et al. 2021

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

Co-saliency detection aims to detect common salient objects from a group of relevant images. Some attempts have been made with the Fully Convolutional Network (FCN) framework and achieve satisfactory detection results. However, due to stacking convolution layers and pooling operation, the boundary details tend to be lost. In addition, existing models often utilize the extracted features without discrimination, leading to redundancy in representation since actually not all features are helpful to the final prediction and some even bring distraction. In this paper, we propose a co-attention module embedded FCN framework, called as Co-Attention FCN (CA-FCN). Specifically, the co-attention module is plugged into the high-level convolution layers of FCN, which can assign larger attention weights on the common salient objects and smaller ones on the background and uncommon distractors to boost final detection performance. Extensive experiments on three popular co-saliency benchmark datasets demonstrate the superiority of the proposed CA-FCN, which outperforms state-of-the-arts in most cases. Besides, the effectiveness of our new co-attention module is also validated with ablation studies.

show abstract

Section: E Comparison With Baselinesmentioning

confidence: 99%

Section: Compared Baselinesmentioning

confidence: 99%

Section: E Comparison With Baselinesmentioning

confidence: 99%

See 1 more Smart Citation

Co-Saliency Detection With Co-Attention Fully Convolutional Network

Gao

Zhao²,

Liu

et al. 2021

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

show abstract

“…Bottom-up object saliency detection Bottom-up approaches [4,5] find objects attracting humans. Different hypothesis or priors [4,5] are used to distinguish salient objects from background, such as global/local contrast [8,21], focusness [22], objectness [7,20,22], cosegment [40] or video motion [19]. These approaches sometimes fail because the hypothesis or priors vary from object category to object category.…”

Section: Related Workmentioning

confidence: 99%

Weakly Supervised Saliency Detection with A Category-Driven Map Generator

Hsu

Lin²,

Chuang

2017

Procedings of the British Machine Vision Conference 2017

Self Cite

View full text Add to dashboard Cite

Top-down saliency detection aims to highlight the regions of a specific object category, and typically relies on pixel-wise annotated training data. In this paper, we address the high cost of collecting such training data by presenting a weakly supervised approach to object saliency detection, where only image-level labels, indicating the presence or absence of a target object in an image, are available. The proposed framework is composed of two deep modules, an image-level classifier and a pixel-level map generator. While the former distinguishes images with objects of interest from the rest, the latter is learned to generate saliency maps so that the training images masked by the maps can be better predicted by the former. In addition to the top-down guidance from class labels, the map generator is derived by also referring to other image information, including the background prior, area balance and spatial consensus. This information greatly regularizes the training process and reduces the risk of overfitting, especially when learning deep models with few training data. In the experiments, we show that our method gets superior results, and even outperforms many strongly supervised methods.

show abstract

“…In other words, in addition to the saliency attribute in an individual image, the repetitiveness constraint across the whole image group is also crucial to suppress the background and noncommon salient regions. In existing methods, the inter-image correspondence is simulated as a matching process [43]- [48], clustering process [49]- [51], low-rank problem [52], [53], propagation process [54]- [56], or learning process [57]- [61]. However, the matching-and propagation-based methods are often time consuming, while the clustering based methods are sensitive to the noise.…”

Section: Introductionmentioning

confidence: 99%

HSCS: Hierarchical Sparsity Based Co-saliency Detection for RGBD Images

Cong

Lei

et al. 2019

IEEE Trans. Multimedia

View full text Add to dashboard Cite

Co-saliency detection aims to discover common and salient objects in an image group containing more than two relevant images. Moreover, depth information has been demonstrated to be effective for many computer vision tasks. In this paper, we propose a novel co-saliency detection method for RGBD images based on hierarchical sparsity reconstruction and energy function refinement. With the assistance of the intra saliency map, the inter-image correspondence is formulated as a hierarchical sparsity reconstruction framework. The global sparsity reconstruction model with a ranking scheme focuses on capturing the global characteristics among the whole image group through a common foreground dictionary. The pairwise sparsity reconstruction model aims to explore the corresponding relationship between pairwise images through a set of pairwise dictionaries. In order to improve the intra-image smoothness and inter-image consistency, an energy function refinement model is proposed, which includes the unary data term, spatial smooth term, and holistic consistency term. Experiments on two RGBD co-saliency detection benchmarks demonstrate that the proposed method outperforms the state-of-the-art algorithms both qualitatively and quantitatively.

show abstract

Segmentation guided local proposal fusion for co-saliency detection

Cited by 12 publications

References 21 publications

Co-Saliency Detection With Co-Attention Fully Convolutional Network

Co-Saliency Detection With Co-Attention Fully Convolutional Network

Weakly Supervised Saliency Detection with A Category-Driven Map Generator

HSCS: Hierarchical Sparsity Based Co-saliency Detection for RGBD Images

Contact Info

Product

Resources

About