2015
DOI: 10.1016/j.jart.2015.07.006
|View full text |Cite
|
Sign up to set email alerts
|

Hexagonal scale invariant feature transform (H-SIFT) for facial feature extraction

Abstract: Feature transformation and key-point identification is the solution to many local feature descriptors. One among such descriptor is the Scale Invariant Feature Transform (SIFT). A small effort has been made for designing a hexagonal sampled SIFT feature descriptor with its applicability in face recognition tasks. Instead of using SIFT on square image coordinates, the proposed work makes use of hexagonal converted image pixels and processing is applied on hexagonal coordinate system. The reason of using the hex… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
17
0
1

Year Published

2017
2017
2022
2022

Publication Types

Select...
4
4
1

Relationship

0
9

Authors

Journals

citations
Cited by 35 publications
(18 citation statements)
references
References 26 publications
0
17
0
1
Order By: Relevance
“…The LFW face database and self-collection database were used to experiment with different improved SIFT models. In this experiment, we compared eleven different models: principal component analysis with SIFT (PCA/SIFT) [50], Gabor/SIFT [51], speeded up robust features with SIFT (SURF/SIFT) [52], partial-descriptor SIFT (PDSIFT) [53], person-specific SIFT [54], SIFT/Kepenekci [55], wavelet transform of the SIFT feature [56], hexagonal SIFT (H-SIFT) [57], and SIFT/RITF. For PCA/SIFT, gradient patches were used around the key points instead of the original patches to make the representation robust to changes in lighting and to reduce the changes that PCA needs to model [50].…”
Section: Experimental Results and Analysismentioning
confidence: 99%
See 1 more Smart Citation
“…The LFW face database and self-collection database were used to experiment with different improved SIFT models. In this experiment, we compared eleven different models: principal component analysis with SIFT (PCA/SIFT) [50], Gabor/SIFT [51], speeded up robust features with SIFT (SURF/SIFT) [52], partial-descriptor SIFT (PDSIFT) [53], person-specific SIFT [54], SIFT/Kepenekci [55], wavelet transform of the SIFT feature [56], hexagonal SIFT (H-SIFT) [57], and SIFT/RITF. For PCA/SIFT, gradient patches were used around the key points instead of the original patches to make the representation robust to changes in lighting and to reduce the changes that PCA needs to model [50].…”
Section: Experimental Results and Analysismentioning
confidence: 99%
“…The method proposed in Reference [56] is based on wavelet transform of the SIFT features. H-SIFT takes advantage of hexagonal transformed image pixels and applies processing on a hexagonal coordinate system, rather than using SIFT on the square image coordinates, displaying better performance [57].…”
Section: Experimental Results and Analysismentioning
confidence: 99%
“…And Face Recognition is Multi-class problem. SVM can be applied to recognize the faces after facial feature extraction [114][115][116][117]or onto the original appearance space. For face recognition, SVM can be applied individually or can be used with the other techniques.…”
Section: Support Vector Machine (Svm)mentioning
confidence: 99%
“…Vùng xung quanh các điểm đặc trưng được xác định và mô tả bằng các véctơ mô tả cục bộ. Véctơ mô tả SIFT (Scale Invariant Feature Transform) [14] được đánh giá rất cao bởi giới chuyên môn trong việc biểu diễn các vùng xung quanh điểm đặc trưng bởi vì nó không đổi đối với những biến đổi tỉ lệ, tịnh tiến, phép quay, và không đổi một phần với đối với những thay đổi về góc nhìn, đồng thời nó cũng rất mạnh với những thay đổi về độ sáng, sự che khuất, nhiễu. Ở bước này, chúng tôi cần tính toán một bộ mô tả ảnh rầy nâu cho một vùng ảnh địa phương sao cho có tính đặc trưng cao (bất biến với các thay đổi khác nhau về độ sáng, thu -phóng ảnh, xoay ảnh,…).…”
Section: B đặC Trưng Cục Bộ Bất Biến Sift Trên ảNhunclassified