SIFT Descriptor for Binary Shape Discrimination, Classification and Matching

Setitra, Insaf; Larabi, Slimane

doi:10.1007/978-3-319-23192-1_41

Cited by 4 publications

(4 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Feature-based matching typically requires more than one feature to describe a complex structure. Such features may include: a Scale Invariant Feature Transform (SIFT) [31], tree union [32], local phase [33], contour feature [34], distance [35], full shape [36], virtual retrieval [37], bag of words [38]. [10], variable-dimensional local shape descriptors (VD-LSD) [10], point feature histogram (PFH) [11], and fast point feature histogram (FPFH) [12].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

AGSF: Adaptive Graph Formulation and Hand-Crafted Graph Spectral Features for Shape Representation

Alwaely

Abhayaratne

2020

IEEE Access

View full text Add to dashboard Cite

Addressing intra-class variation in high similarity shapes is a challenging task in shape representation due to highly common local and global shape characteristics. Therefore, this paper proposes a new set of hand-crafted features for shape recognition by exploiting spectral features of the underlying graph adaptive connectivity formed by the shape characteristics. To achieve this, the paper proposes a new method for formulating an adaptively connected graph on the nodes of the shape outline. The adaptively connected graph is analysed in terms of its spectral bases followed by extracting hand-crafted adaptive graph spectral features (AGSF) to represent both global and local characteristics of the shape. Experimental evaluation using five 2D shape datasets and four challenging 3D shape datasets shows improvements with respect to the existing hand-crafted feature methods up to 9.14% for 2D shapes and up to 14.02% for 3D shapes. Also for 2D datasets, the proposed AGSF has outperformed the deep learning methods by 17.3%.INDEX TERMS Shape recognition, graph signal processing, graph spectral theory, adaptive graph connectivity, hand-crafted features, graph spectral feature learning.

show abstract

Section: Related Workmentioning

confidence: 99%

“…[10], variable-dimensional local shape descriptors (VD-LSD) [10], point feature histogram (PFH) [11], and fast point feature histogram (FPFH) [12]. The classification process is usually implemented using the Bayesian classifier [31], the Nearest Neighbors (NN) [33] or Support Vector Machines (SVM) [36].…”

Section: Related Workmentioning

confidence: 99%

AGSF: Adaptive Graph Formulation and Hand-Crafted Graph Spectral Features for Shape Representation

Alwaely

Abhayaratne

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…This approach has been used extensively to extract features from data collections where it is known in advance that the shape of interest occupies the majority of the image area, such as on the MNIST [74] and OMINGLOT [72] databases of handwritten characters. Other features based on pixel information are Haar features [114], ring projection [130], shape context [10] and SIFT key points [77] applied for greyscale graphics [106], the ImageNet dataset [60] and the ''Tarragona'' image repository [86].…”

Section: Extraction Of Featuresmentioning

confidence: 99%

New trends on digitisation of complex engineering drawings

Moreno‐García

Elyan

Jayne

2018

Neural Comput & Applic

View full text Add to dashboard Cite

Engineering drawings are commonly used across different industries such as oil and gas, mechanical engineering and others. Digitising these drawings is becoming increasingly important. This is mainly due to the legacy of drawings and documents that may provide rich source of information for industries. Analysing these drawings often requires applying a set of digital image processing methods to detect and classify symbols and other components. Despite the recent significant advances in image processing, and in particular in deep neural networks, automatic analysis and processing of these engineering drawings is still far from being complete. This paper presents a general framework for complex engineering drawing digitisation. A thorough and critical review of relevant literature, methods and algorithms in machine learning and machine vision is presented. Real-life industrial scenario on how to contextualise the digitised information from specific type of these drawings, namely piping and instrumentation diagrams, is discussed in details. A discussion of how new trends on machine vision such as deep learning could be applied to this domain is presented with conclusions and suggestions for future research directions.

show abstract

“…Shapes are also recognized using feature-based representation that typically requires more than one feature to describe a complex structure. Such features may include a scale invariant feature transform (SIFT) [13], tree union [3], local phase [14], distance and the central point [4], contour features and distance [15]. With regards to the graph-based studies, using graph for 2D shape matching [16], [17] mainly relied on bipartite matching.…”

Section: Introductionmentioning

confidence: 99%

Graph Spectral Domain Shape Representation

Alwaely

Abhayaratne

2018

2018 26th European Signal Processing Conference (EUSIPCO)

View full text Add to dashboard Cite

One of the major challenges in shape matching is recognising and interpreting the small variations in objects that are distinctly similar in their global structure, as in well known ETU10 silhouette dataset and the Tool dataset. The solution lies in modelling these variations with numerous precise details. This paper presents a novel approach based on fitting shape's local details into an adaptive spectral graph domain features. The proposed framework constructs an adaptive graph model on the boundaries of silhouette images based on threshold, in such a way that reveals small differences. This follows feature extraction on the spectral domain for shape representation. The proposed method shows that interpreting local details leading to improve the accuracy levels by 2% to 7% for the two datasets mentioned above, respectively.

show abstract

SIFT Descriptor for Binary Shape Discrimination, Classification and Matching

Cited by 4 publications

References 18 publications

AGSF: Adaptive Graph Formulation and Hand-Crafted Graph Spectral Features for Shape Representation

AGSF: Adaptive Graph Formulation and Hand-Crafted Graph Spectral Features for Shape Representation

New trends on digitisation of complex engineering drawings

Graph Spectral Domain Shape Representation

Contact Info

Product

Resources

About