Search

Shape characterization plays a highly prominent role in retrieval and relies extremely upon descriptors inbuilt with lightweight operations and compaction qualities. However, realizing such a simple and robust shape descriptor capable of dealing with noise, variations in brightness, and deformations pose a significant challenge. In this regard, a simple and effective shape descriptor using the Quadratic Bezier Triangle (QBT) targeting shape matching and retrieval is presented in this work. The mechanism commences with triangular tessellation of each image followed by determining their side-wise intensity differences that are then mapped to QBT vertices to yield the order-wise control points. The maxima of the resulting points are transformed into the binary equivalent of the given shape that is compacted into octal values. Later, these localized values are globally transformed into shape histograms to yield the QBT-based Feature (QBTF) descriptor representing the given input. QBTFs performance is exhaustively analyzed in terms of Bulls Eye Retrieval (BER) score and classification accuracy using the public datasets namely Kimia-99, MPEG-7 CE-1 part B, PHOS, and Tari-1000. Relative BER investigations witnessed across diverse datasets reveal a consistent and improved score of 93% achieved by this scheme over its peers. Mathematical analysis of invariance, noise resilience, and the complexities involved in realizing QBTF, establish the robustness and suitability of this scheme for real-time shape description. • Quadratic Bezier Triangle for shape characterization is the first of its kind and is the core novelty of this contribution. • The engaged Bernstein polynomials acutely capitulate high-frequency features deemed essential for shape characterization. • Inherent convex nature of QBT demonstrates high robustness towards diverse affine transformations, noise, and illumination. • Rigorous analysis performed on benchmark shape datasets reveals the intention's superiority. • The proposed shape descriptor's performance matches with black-box models at a relatively lesser computational cost. [ABSTRACT FROM AUTHOR]

We introduce a new concept for detecting the saliency of 3-D shapes, that is, human-centered saliency (HCS) detection on the surface of shapes, whereby a given shape is analyzed not based on geometric or topological features directly obtained from the shape itself, but by studying how a human uses the object. Using virtual agents to simulate the ways in which humans interact with objects helps to understand shapes and detect their salient parts in relation to their functions. HCS detection is less affected by inconsistencies between the geometry or topology of the analyzed 3-D shapes. The potential benefit of the proposed method is that it is adaptable to variable shapes with the same semantics, as well as being robust against a geometrical and topological noise. Given a 3-D shape, its salient part is detected by automatically selecting a corresponding agent and making them interact with each other. Their adaption and alignment depend on an optimization framework and a training process. We demonstrate the detected salient parts for different types of objects together with the stability thereof. The salient parts can be used for important vision tasks, such as 3-D shape retrieval. [ABSTRACT FROM PUBLISHER]

Measuring the similarity between two instances reliably, shape or image, is a challenging problem in shape and image retrieval. In this paper, a simple yet effective method called Neighbor Set Similarity (NSS) is proposed, which is superior to both traditional pairwise similarity and diffusion process. NSS makes full use of contextual information to capture the geometry of the underlying manifold, and obtains a more precise measure than the original pairwise similarity. Moreover, based on NSS, we propose a powerful fusion process to utilize the complementarity of different descriptors to further enhance the retrieval performance. The experimental results on MPEG-7 shape dataset, N-S image dataset and ORL face dataset demonstrate the effectiveness of the proposed method. In addition, the time complexity of NSS is much lower than diffusion process, which suggests that NSS is more suitable for large scale image retrieval than diffusion process. [ABSTRACT FROM AUTHOR]

Abstract: Manifold learning is a non-linear method with the aim of finding a constructive way to embed the data from a high-dimensional space into a low-dimensional one. The improvement of shape classification is the major approach in this paper which is based on the continuity in feature space in accordance with the continuity in semantic one. In this regard, a non-linear approach is employed to map the shape feature vectors to a new space while their semantics become similar with human opinion, the Euclidean distance between two feature vectors would be closed. Shapes are described by four contour-based and region-based techniques by the proposed method. In other word, they are described in four observation spaces. Furthermore, a parameter space is learnt from multiple observation spaces based on fusion of dissimilarities in a supervised manner. Experimental results show the validity and efficiency of the proposed approach for shape classification over a variety of standard shape datasets. [Copyright &y& Elsevier]

Non-grid shape retrieval has been an important and challenging task that usually requires high-level and discriminative shape features. However, most of the existing descriptors are hand-crafted, which fail to distinguish and focus on important features from different regions and scales for given query shape. To tackle this issue, this paper presents a novel shape retrieval method, namely BoF-DM (Bag of feature with Discriminative Module), which is specially designed to highlight discriminative features from the multi-scale context structures. Specifically, a shape decomposing method is designed by combining the KPs (key points) and ring graph, result in local to global feature learning. To obtain high-level shape features, the spatial context of each contour fragment is learned by the stacked lightweight network in an unsupervised manner. In people's cognition, discriminative parts, such as heads or legs, play an essential role for identifying animal shape classes. Inspired by this phenomenon, we propose a TF-IDF (term frequency-inverse document frequency) with entropy-based Discriminative Module, which fully considers the attention issue of inter-class difference and intra-class similarity. Finally, decision-making is done by metric similarity. The experiments demonstrate that the proposed method achieves the high BER (Bull's Eye Retrieval) rates of 93.63%, 100%, 100%, and 99.00% on the considered MPEG-7, Kimia's 99, Kimia's 216, and Articulated shape datasets respectively. Furthermore, our method also obtains high performance under noisy and articulated conditions. [ABSTRACT FROM AUTHOR]