Shape-tailored local descriptors and their application to segmentation and tracking

Khan, Naeemullah; Algarni, Marei; Yezzi, Anthony; Sundaramoorthi, Ganesh

doi:10.1109/cvpr.2015.7299014

Cited by 9 publications

(23 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We test on two datasets used in [23]. The Brodatz Synthetic Dataset has 198 images generated from textures in Brodatz and random shapes from MPEG dataset.…”

Section: Texture Segmentation Datasets and Methods Comparedmentioning

confidence: 99%

“…We use RGB color channels and binned oriented gradients at four angles, as the features for segmentation. Since the contribution in this paper is the use of shape-tailored scale spaces at a continuum of scales, we compare to [23] (STLD), which uses scale space but only considers a discrete number of scales. For reference, we include other segmentation methods.…”

Section: Texture Segmentation Datasets and Methods Comparedmentioning

confidence: 99%

“…Anisotropic scale spaces [40,4] have been applied to reduce blurring across boundaries, but this could blur across regions where edges are not salient. Recently, [23] have addressed this issue by computing scales locally within the evolving regions of the segmentation. However, only a discrete number of scales are used and thus the method does not exhibit coarse-to-fine behavior.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Coarse-to-Fine Segmentation with Shape-Tailored Continuum Scale Spaces

Khan

Hong

Yezzi

et al. 2017

2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

Self Cite

View full text Add to dashboard Cite

“…We test on two datasets used in [23]. The Brodatz Synthetic Dataset has 198 images generated from textures in Brodatz and random shapes from MPEG dataset.…”

Section: Texture Segmentation Datasets and Methods Comparedmentioning

confidence: 99%

Section: Texture Segmentation Datasets and Methods Comparedmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Coarse-to-Fine Segmentation with Shape-Tailored Continuum Scale Spaces

Khan

Hong

Yezzi

et al. 2017

2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

Self Cite

View full text Add to dashboard Cite

“…For this reason, this paper employs the shape-specific feature (SSF) [5], which is tolerant to multicolor rooftops, sensitive to various shapes, and efficient in computation. In detail, SSF extraction consists of the following three steps.…”

Section: Shape-specific Feature Extractionmentioning

confidence: 99%

“…Shape-specific feature is extracted using the method proposed by Khan et al [5] for capturing complex shapes and structures of buildings. 2.…”

Section: Introductionmentioning

confidence: 99%

Utilizing Shape-Based Feature and Discriminative Learning for Building Detection

Zhang

Shen

Huo

2017

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYBuilding detection from high resolution remote sensing images is challenging due to the high intraclass variability and the difficulty in describing buildings. To address the above difficulties, a novel approach is proposed based on the combination of shape-specific feature extraction and discriminative feature classification. Shape-specific feature can capture complex shapes and structures of buildings. Discriminative feature classification is effective in reflecting similarities among buildings and differences between buildings and backgrounds. Experiments demonstrate the effectiveness of the proposed approach.

show abstract

Have We Solved Edge Detection? A Review of State-of-the-Art Datasets and DNN Based Techniques

et al. 2022

Self Cite

View full text Add to dashboard Cite

Recent Deep Neural Networks (DNNs) based edge detection methods claim to beat human performance on small scale datasets like BSDS500. But is the problem of edge detection really solved? To answer this question, we review the existing dataset limitations as well as the generalization capabilities of the proposed architectures. To this end, we develop a Synthetic Textured Masks Dataset (STMD) that contains 28,000 gray scale images. The performance of several edge detection methods is severely degraded on STMD. To further validate these results we propose a baseline Single Scale Feed Forward Edge Detector (SFED), which is a simple 9-layer feed-forward convolutional neural network with no pooling layers. The performance of SFED is better than most state-of-the-art architectures on BSDS500 and is superior to all the compared architectures on STMD. These results show that most of the architectural advancements of existing architectures are at the cost of generalizability where if we change the dataset set distribution (both training and testset), the performance become significantly degraded and therefore the problem of edge detection is still far away from being solved. There are also severe limitations of existing datasets in the field, and STMD provides a framework for designing and testing better edge detection architectures for novel application areas, such as, medical imaging and self-driving cars.

show abstract

Shape-tailored local descriptors and their application to segmentation and tracking

Cited by 9 publications

References 46 publications

Coarse-to-Fine Segmentation with Shape-Tailored Continuum Scale Spaces

Coarse-to-Fine Segmentation with Shape-Tailored Continuum Scale Spaces

Utilizing Shape-Based Feature and Discriminative Learning for Building Detection

Have We Solved Edge Detection? A Review of State-of-the-Art Datasets and DNN Based Techniques

Contact Info

Product

Resources

About