Point Context: An Effective Shape Descriptor for RST-Invariant Trajectory Recognition

Motion trajectory recognition is one of the most important means to determine the identity of a moving object. A compact and discriminative feature representation method can improve the trajectory recognition accuracy. This paper presents an efficient framework for action recognition using a three-dimensional skeleton kinematic joint model. First, we put forward a rotation-scale-translation-invariant shape descriptor based on point context (PC) and the normal vector of hypersurface to jointly characterize local motion and shape information. Meanwhile, an algorithm for extracting the key trajectory based on the confidence coefficient is proposed to reduce the randomness and computational complexity. Second, to decrease the eigenvalue decomposition time complexity, a tensor shape descriptor (TSD) based on PC that can globally capture the spatial layout and temporal order to preserve the spatial information of each frame is proposed. Then, a multilinear projection process is achieved by tensor dynamic time warping to map the TSD to a low-dimensional tensor subspace of the same size. Experimental results show that the proposed shape descriptor is effective and feasible, and the proposed approach obtains considerable performance improvement over the state-of-the-art approaches with respect to accuracy on a public action dataset.

Dimensionality reduction is of great importance which aims at reducing the spectral dimensionality while keeping the desirable intrinsic structure information of hyperspectral images. Tensor analysis which can retain both spatial and spectral information of hyperspectral images has caused more and more concern in the field of hyperspectral images processing. In general, a desirable low dimensionality feature representation should be discriminative and compact. To achieve this, a tensor discriminant analysis model via compact feature representation (TDA-CFR) was proposed in this paper. In TDA-CFR, the traditional linear discriminant analysis was extended to tensor space to make the resulting feature representation more informative and discriminative. Furthermore, TDA-CFR redefines the feature representation of each spectral band by employing the tensor low rank decomposition framework which leads to a more compact representation.

Compact feature representation of person image is important for person re-identification (Re-ID) task. Recently, part-based representation models have been widely studied for extracting the more compact and robust feature representation for person image to improve person Re-ID results. However, existing part-based representation models mostly extract the features of different parts independently which ignore the spatial relationship information among different parts. To address this issue, in this paper we propose a novel deep learning framework, named Part-based Hierarchical Graph Convolutional Network (PH-GCN) for person Re-ID problem. Given a person image, PH-GCN first constructs a hierarchical graph to represent the spatial relationships among different parts. Then, both local and global feature learning is achieved by the feature information passing in PH-GCN, which takes the information of other parts into account for part feature representation. Finally, a perceptron layer is adopted for the final person part label prediction and re-identification. The proposed framework provides a general solution that integrates <i>local</i>, <i>global</i> and <i>structural</i> feature learning simultaneously in a unified end-to-end network representation and learning. Extensive experiments on several widely used benchmark datasets demonstrate the effectiveness and benefits of the proposed PH-GCN approach for person Re-ID task.

Multiview data, characterized by rich features, are crucial in many machine learning applications. However, effectively extracting intraview features and integrating interview information present significant challenges in multiview learning (MVL). Traditional deep network-based approaches often involve learning multiple layers to derive latent. In these methods, the features of different classes are typically implicitly embedded rather than systematically organized. This lack of structure makes it challenging to explicitly map classes to independent principal subspaces in the feature space, potentially causing class overlap and confusion. Consequently, the capability of these representations to accurately capture the intrinsic structure of the data remains uncertain. In this article, we introduce an innovative multiview representation learning (MVRL) by maximizing two information-theoretic metrics: intraview coding rate reduction and interview mutual information. Specifically, in the intraview representation learning, we aim to optimize feature representations by maximizing the coding rate difference between the entire dataset and individual classes. This process expands the feature representation space while compressing the representations within each class, resulting in more compact feature representations within each viewpoint. Subsequently, we align and fuse these view-specific features through space transformation and cross-sample fusion to achieve consistent representation across multiple views. Finally, we maximize information transmission to maintain consistency and correlation among data representations across views. By maximizing mutual information between the consensus representations and view-specific representations, our method ensures that the learned representations capture more concise intrinsic features and correlations among different views, thereby enhancing the performance and generalization ability of MVL. Experiments show that the proposed methods have achieved excellent performance.

Semi-supervised learning (SSL), as one of the dominant methods, aims at leveraging the unlabeled data to deal with the annotation dilemma of supervised learning, which has attracted much attentions in the medical image segmentation. Most of the existing approaches leverage a unitary network by convolutional neural networks (CNNs) with compulsory consistency of the predictions through small perturbations applied to inputs or models. The penalties of such a learning paradigm are that (1) CNN-based models place severe limitations on global learning; (2) rich and diverse class-level distributions are inhibited. In this paper, we present a novel CNN-Transformer learning framework in the manifold space for semi-supervised medical image segmentation. First, at intra-student level, we propose a novel class-wise consistency loss to facilitate the learning of both discriminative and compact target feature representations. Then, at inter-student level, we align the CNN and Transformer features using a prototype-based optimal transport method. Extensive experiments show that our method outperforms previous state-of-the-art methods on three public medical image segmentation benchmarks.

Using unlabeled data in a sparse-coding framework for human activity recognition

Open-set face identification with index-of-max hashing by learning

IntroductionFungal identification through ITS sequencing is pivotal for biodiversity and ecological studies, yet existing methods often face challenges with high-dimensional features and inconsistent taxonomy predictions.MethodWe proposed HFTC, a hierarchical fungal taxonomic classifier built upon a multi-level random forest (RF) architecture. Notably, HFTC incorporates a bidirectional k-mer strategy to capture contextual information from both sequence orientations. By leveraging Word2Vec embedding, it reduces feature dimensionality from 4k to only 200, significantly improving computational efficiency while preserving rich sequence context.ResultExperimental results demonstrate that HFTC outperforms Mothur, RDP, Sintax, QIIME2, and CNN-Duong, achieving a Matthews correlation coefficient (MCC) of 95.31% despite uneven class distributions. Its overall accuracy (ACC) reaches 95.25%. At the species level, it attains a hierarchical accuracy (HA) of 95.10%, surpassing the best-performing deep learning baseline, CNN-Duong, by 3.2%. Moreover, HFTC exhibits the smallest discrepancy between ACC and HA (1.60%), in contrast to CNN-Duong, which shows the largest gap (35.00%), highlighting HFTC’s superior hierarchical consistency.DiscussionHFTC offers a scalable and accurate approach for fungal taxonomic classification. Its compact feature representation and hierarchical architecture make it particularly suitable for microbial diversity research. The source code and datasets are publicly accessible at https://github.com/wjjw0731/HFTC/tree/master.

Recently, the latest advances in compact feature representation and feature learning have provided an efficient framework for several visual analysis tasks, such as object recognition. However, when multiple cameras with overlapping fields-of-view are employed, other visual analysis tasks such as depth estimation can be supported and object recognition accuracy can be improved. In this paper the problem of distributed visual analysis from multiple views of a scene is addressed, considering that computational power and bandwidth, at each camera sensor, are rather limited. More specifically, an efficient coding technique for local binary features is proposed which exploits the correlation at the decoder side between each descriptor and its quantized representation. Moreover, considering that descriptors representing the same visual feature across different views are well correlated, a technique to avoid the transmission of redundant descriptors from multiple views is proposed. At the decoder, the joint statistics of all descriptors from all views is used to drive the selection of the best descriptors to be transmitted by each sensing node. The proposed multi-view feature coding and selection techniques allow obtaining bitrate reductions up to 80%, with respect to the uncompressed descriptor rate, for a certain task accuracy.

Electrocardiogram (ECG) signals are crucial for monitoring cardiac activity and diagnosing various cardiovascular conditions. However, these signals are often contaminated by different types of noise, such as baseline wander, muscle artifacts, and power line interference, which can obscure critical information and hinder accurate diagnosis. This study used a 1-Dimensional Convolutional Neural Network (1D CNN) architecture with seven convolutional layers for denoising ECG signals. The model utilizes a fully convolutional autoencoder approach, comprising an encoder that transforms noisy input signals into compact feature representations and a decoder that reconstructs the cleaned signals. The proposed architecture was tested using the MIT-BIH Noise Stress Test Database, which includes ECG recordings with simulated noise conditions. Performance evaluation metrics such as Mean Squared Error (MSE), Signal-to-Noise Ratio (SNR), and Mean Absolute Deviation (MAD) were used to assess the model's effectiveness. Results showed a low MSE of 0.034, a high SNR of 15.8 dB, and a MAD of 0.754, indicating significant noise reduction and high-quality signal reconstruction. These findings demonstrate that the 1D CNN architecture effectively reduces various types of noise in ECG signals, thereby enhancing signal quality and facilitating more accurate analysis and diagnosis. The model's ability to maintain the integrity of crucial ECG features while removing noise suggests its potential utility in clinical applications for improving cardiovascular disease diagnosis

Maji and Berg have recently introduced an explicit feature map approximating the intersection kernel. This enables efficient learning methods for linear kernels to be applied to the non-linear intersection kernel, expanding the applicability of this model to much larger problems. In this paper we generalize this idea, and analyse a large family of additive kernels, called homogeneous, in a unified framework. The family includes the intersection, Hellinger's, and χ <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> kernels commonly employed in computer vision. Using the framework we are able to: (i) provide explicit feature maps for all homogeneous additive kernels along with closed form expression for all common kernels; (ii) derive corresponding approximate finite-dimensional feature maps based on the Fourier sampling theorem; and (iii) quantify the extent of the approximation. We demonstrate that the approximations have indistinguishable performance from the full kernel on a number of standard datasets, yet greatly reduce the train/test times of SVM implementations. We show that the χ <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> kernel, which has been found to yield the best performance in most applications, also has the most compact feature representation. Given these train/test advantages we are able to obtain a significant performance improvement over current state of the art results based on the intersection kernel.

This paper presents an effective local spatial-temporal descriptor for action recognition from skeleton sequences. The unique property of our descriptor is that it takes the spatial-temporal discrimination and action speed variations into account, intending to solve the problems of distinguishing similar actions and identifying actions with different speeds in one goal. The entire algorithm consists of two stages. First, a frame selection method is used to remove noisy skeletons for a given skeleton sequence. From the selected skeletons, skeleton joints are mapped to a high dimensional space, where each point refers to kinematics, time label and joint label of a skeleton joint. To encode relative relationships among joints, pairwise points from the space are then jointly mapped to a new space, where each point encodes the relative relationships of skeleton joints. Second, Fisher Vector (FV) is employed to encode all points from the new space as a compact feature representation. To cope with speed variations in actions, an energy-based temporal pyramid is applied to form a multi-temporal FV representation, which is fed into a kernel-based extreme learning machine classifier for recognition. Extensive experiments on benchmark datasets consistently show that our method outperforms state-of-the-art approaches for skeleton-based action recognition.

With the increasing number of available digital images, there is an urgent need of image content description to facilitate content based image retrieval (CBIR). Besides colour and texture, shape is an important low level feature in describing image content. An object can be photographed from different distances and angles. However, we often want to classify the images of the same object into one class, despite the change of perspective. So, it is desired to extract shape features that are invariant to the change of perspective. The shape of an object from one viewpoint to another can be linked through an affine transformation, if it is viewed from a much larger distance than its size along the line of sight. Those invariant shape features are known as affine invariant shape representations. Because of the change of perspective, it is more difficult to develop affine invariant shape representations than normal ones. The goal of this work is to develop affine invariant shape descriptors. Through shape retrieval experiments, we find that the performance of the existing affine invariant shape representations are not satisfactory. Especially, when the shape boundary is corrupted by noise, their performance degrades quickly. In this work, two new affine invariant contour-based shape descriptors, the ICA Fourier shape descriptor (ICAFSD) and the whitening Fourier shape descriptor (WFSD) have been developed. They perform better than most of the existing affine invariant shape representations, while having compact feature size and low computational time requirement. Four region-based affine-invariant shape descriptors, the ICA Zernike moment shape descriptor (ICAZMSD), the whitening Zernike moment shape descriptor (WZMSD), the ICA orthogonal Fourier Mellin moment shape descriptor (ICAOFMMSD), and the whitening orthogonal Fourier Mellin moment shape descriptor (WOFMMSD), are also proposed, in this work. They can be applied to both simple and complex shapes, and have close to perfect performance in retrieval experiments. The advantage of those newly proposed shape descriptors is even more apparent in experiments on shapes with added boundary noise: Their performance does not deteriorate as much as the existing ones.

With the increasing number of available digital images, there is an urgent need of image content description to facilitate content based image retrieval (CBIR). Besides colour and texture, shape is an important low level feature in describing image content. An object can be photographed from different distances and angles. However, we often want to classify the images of the same object into one class, despite the change of perspective. So, it is desired to extract shape features that are invariant to the change of perspective. The shape of an object from one viewpoint to another can be linked through an affine transformation, if it is viewed from a much larger distance than its size along the line of sight. Those invariant shape features are known as affine invariant shape representations. Because of the change of perspective, it is more difficult to develop affine invariant shape representations than normal ones. The goal of this work is to develop affine invariant shape descriptors. Through shape retrieval experiments, we find that the performance of the existing affine invariant shape representations are not satisfactory. Especially, when the shape boundary is corrupted by noise, their performance degrades quickly. In this work, two new affine invariant contour-based shape descriptors, the ICA Fourier shape descriptor (ICAFSD) and the whitening Fourier shape descriptor (WFSD) have been developed. They perform better than most of the existing affine invariant shape representations, while having compact feature size and low computational time requirement. Four region-based affine-invariant shape descriptors, the ICA Zernike moment shape descriptor (ICAZMSD), the whitening Zernike moment shape descriptor (WZMSD), the ICA orthogonal Fourier Mellin moment shape descriptor (ICAOFMMSD), and the whitening orthogonal Fourier Mellin moment shape descriptor (WOFMMSD), are also proposed, in this work. They can be applied to both simple and complex shapes, and have close to perfect performance in retrieval experiments. The advantage of those newly proposed shape descriptors is even more apparent in experiments on shapes with added boundary noise: Their performance does not deteriorate as much as the existing ones.

Recently, unsupervised Hashing has attracted much attention in the machine learning and information retrieval communities, due to its low storage and high search efficiency. Most of existing unsupervised Hashing methods rely on the local semantic structure of the data as the guiding information, requiring to preserve such semantic structure in the Hamming space. Thus, how to precisely represent the local structure of the data and Hashing code s becomes the key point to success. This study proposes a novel Hashing method based on self-supervised learning. Specifically, it is proposed to utilize the contrast learning to acquire a compact and accurate feature representation for each sample, and then a semantic structure matrix can be constructed for representing the similarity between samples. Meanwhile, a new loss function is proposed to preserve the semantic information and improve the discriminative ability in the Hamming space, by the spirit of the instance discrimination method proposed recently. The proposed framework is end-to-end trainable. Extensive experiments on two large-scale image retrieval data sets show that the proposed method can significantly outperform current state-of-the-art methods.