Design and perceptual validation of performance measures for salient object segmentation

By Vida Movahedi, James H. Elder. 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Workshops, 2010

473 citations recorded by the source corpus.

Subfields: Cognitive Neuroscience, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/283743048be918a1a4144d78ea65f45398c37e2d

How researchers see this paper

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

In Cognitive Neuroscience

In Computer Vision and Pattern Recognition

Citations in the corpus

Salient Object Detection in the Deep Learning Era: An In-Depth Survey (2019, methodology): “We construct a hybrid benchmark consists of 1,800 distinctive images randomly selected from 6 datasets (300 each), namely SOD [118], ECSSD [34], DUT-OMRON [52], PASCAL-S [59], HKU-IS [26] and the test set of DUTS [73].”
Salient Object Detection via Structured Matrix Decomposition (2017, methodology): “In particular, we use MSRA10K [69] and DUTOMRON [61] for images with a single salient object, iCoSeg [71] and SOD [70] for cases with multiple salient objects, and ECSSD [21] for images with complex scenes.”
Advanced Deep-Learning Techniques for Salient and Category-Specific Object Detection: A Survey (2018, methodology): “The ECSSD data set [81] has 1,000 images with some overlapped images in the SOD data set [82].”
PiCANet: Learning Pixel-Wise Contextual Attention for Saliency Detection (2017, background): “SOD [29] contains 300 images with complex backgrounds and multiple foreground objects.”
Global Context-Aware Progressive Aggregation Network for Salient Object Detection (2020, background): “• SOD (Movahedi and Elder 2010) is composed of 300 images, many of which contain multiple objects either with low contrast or touching the image boundary;”
Geodesic Saliency Using Background Priors (2012, methodology): “Distribution of percentage of background pixels along the image boundary on the MSRA-1000 database [9] (left) and Berkeley-300 database [30] (right).”
Object Detection With Deep Learning: A Review (2018, methodology): “The evaluated approaches include R-CNN [15], SPP-net [64], Fast R-CNN [16], networks on convolutional feature maps (NOC) [114], Bayes [85], MR-CNN& S-CNN [105], Faster R-CNN [17], HyperNet [101], ION [95], MS-GR [104], StuffNet [100], SSD300 [71], SSD512 [71], OHEM [113], SDP + CRC [34], G-CNN [70],”
What Makes a Patch Distinct? (2013, unknown): “SOD [17]: 300 images taken from the Berkeley Segmentation Dataset for which seven users selected the boundaries of the salient objects.”
DeepSaliency: Multi-Task Deep Neural Network Model for Salient Object Detection (2015, methodology): “1) Datasets: In order to evaluate the performance of the proposed approach, we conduct a set of qualitative and quantitative experiments on eight benchmark datasets annotated with pixel-wise ground-truth labeling, including the ASD [29], DUT-OMRON [17], ECSSD [20], PASCAL-S [48], THUR [49], THUS [35”
EGNet: Edge Guidance Network for Salient Object Detection (2019, unknown): “The goal of salient object detection (SOD) is to ﬁnd the most visually distinctive objects in an image.”
BASNet: Boundary-Aware Salient Object Detection (2019, methodology): “Particularly, our method improves relaxF bβ by 4.1%, 5.1%, 6.2%, 6.2%, 3.4%, 5.9% on SOD, ECSSD, DUT-OMRON, PASCAL-S, HKU-IS and DUTS-TE datasets, respectively.”
Deep networks for saliency detection via local estimation and global search (2015, methodology): “We evaluate the proposed algorithm on four benchmark data sets: MSRA-5000 [24], SOD [27], ECCSD [40] and PASCAL-S [23].”
Contextual Hypergraph Modeling for Salient Object Detection (2013, background): “SOD [28] is composed of 300 challenging images.”
Salient Object Detection With Pyramid Attention and Salient Edges (2019, methodology): “We conduct extensive experiments on six popular benchmarks: ECCSD [49], DUT-OMRON [50], HKU-IS [21], PASCAL-S [25], SOD [30], and DUTS-TE [35], which are all publicly available and are human-labeled with pixel-wise ground truth for quantitative evaluations.”
U2-Net: Going Deeper with Nested U-Structure for Salient Object Detection (2020, methodology): “There is a common pattern in the design of most SOD networks [18, 27, 41, 6], that is, they focus on making good use of deep features extracted by existing backbones, such as Alexnet [17], VGG [35], ResNet [12], ResNeXt [44], DenseNet [15], etc.”
Learning Video Object Segmentation from Static Images (2016, methodology): “We consider images and segmentation masks from ECSSD [41], MSRA10K [9], SOD [28], and PASCAL-S [25].”