Set Of Local Features Research Articles

Spatio-temporal trajectory data modeling for fishing gear classification

International Organizations urge the protection of our oceans and their ecosystems due to their immeasurable importance to humankind. Since illegal fishing activities, commonly known as IUU fishing, cause irreparable damage to these ecosystems, concerned organisms are pushing to detect and combat IUU fishing practices. The automatic identification system allows to locate the position and trajectory of fishing vessels. In this study we address the task of detecting vessels’ fishing gears based on the trajectory behavior defined by GPS position data, a useful task to prevent the proliferation of IUU fishing practices. We present a new database including trajectories that span 7 different fishing gears and analyze these as in a time sequence analysis problem. We leverage from feature extraction techniques from the online signature verification domain to model vessel trajectories, and extract relevant information in the form of both local and global feature sets. We show how, based on these sets of features, the kinematics of vessels according to different fishing gears can be effectively classified using common supervised learning algorithms with accuracies up to 90%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$90\\%$$\\end{document}. Furthermore, motivated by the concerns raised by several organizations on the adverse impact of bottom trawling on marine biodiversity, we present a binary classification experiment in which we were able to distinguish this kind of fishing gear with an accuracy of 99%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$99\\%$$\\end{document}. We also illustrate in an ablation study the relevance of factors such as data availability and the sampling period to perform fishing gear classification. Compared to existing works, we highlight these factors, especially the importance of using sampling periods in the order of minutes instead of hours.

Online electronic signature recognition using sparse classification techniques that support neural models

With the rapid development of information technology, electronic signature plays an increasingly important role in people’s production practice. However, there are a large number of hackers maliciously stealing information in the network. In order to avoid this phenomenon, we urgently need to strengthen the research on online electronic signature recognition technology. Based on the sparse classification technology of neural model, this paper constructs an online electronic signature recognition model by using convolutional neural network and sparse classification technology. We first extract the local features of online electronic signatures, construct feature vectors and perform sparse representation. Sub-model we construct a scheme for online electronic signature recognition based on neural models and sparse classification techniques using a combination of algorithms. We first extract the local features of online electronic signatures, construct feature vectors and perform sparse representation. At the same time, the features in the training image set are extracted, local feature sets are constructed, feature dictionaries are created, and the vectors in the feature dictionaries are matched with the global sparse vectors constructed by the electronic signatures to be detected, and the matching results are finally obtained. At the same time, the features in the training image set are extracted, the local feature set is constructed, the feature dictionary is created, and the vector in the feature dictionary is matched with the global sparse vector constructed by the electronic signature to be detected, and finally the matching result is obtained. In order to verify the accuracy of the model, we first extracted 1000 respondents for online e-signature recognition experimental results show that the recognition accuracy of online e-signature has been significantly improved. Finally, in order to determine the optimal number of training sets for the model constructed in this experiment, we analyzed the correlation between training and sample size and recognition accuracy. Finally, it was concluded that the recognition accuracy increased with the increase of the number of training samples. Electronic signatures can quickly examine the signature results, and electronic signature recognition can be used to fix and tamper-proof evidence to enhance the security and trustworthiness of signatures, and it is imperative to improve the security of electronic signatures. In this paper, we study online electronic signature recognition technology, using neural model and sparse classification to construct an efficient and accurate recognition model. Experiments show that the model is effective and the number of training samples affects the recognition accuracy. This paper provides a new approach for the development of this technique. When the training samples are greater than 1300, the recognition accuracy is stable at 95%. This research has certain theoretical and practical significance, and promotes the rapid development of online electronic signature recognition.

Structure correspondence searching of CAD model using local feature-based description and indexing

The CAD model retrieval has played a significant role in various applications, including product development and knowledge mining. However, most existing retrieval methods compare 3D shape similarity from a global perspective, while detecting similar structures automatically for CAD models remains a challenging problem. Consequently, this study proposes a structure correspondence searching framework for CAD models to address the issues. According to the boundary representation (B-rep) information, the proposed method first segments a CAD model into a set of local features denoted as structural cells. Then, the descriptor of each structural cell is extracted using a weighted shape distribution vector and neighbor set. In order to speed up the matching of structural cells, an indexing and filtering mechanism is constructed based on the shape clustering and topological analysis. The matched structural cells determine the boundary of similar structures. Finally, similarity measurement is conducted to generate a ranking list by analyzing the quality of the matched structural cells. The rationality and efficiency of the proposed approach are demonstrated via an analysis of experimental results.

Low-Data Deep Quantum Chemical Learning for Accurate MP2 and Coupled-Cluster Correlations.

Accurate ab initio prediction of electronic energies is very expensive for macromolecules by explicitly solving post-Hartree-Fock equations. We here exploit the physically justified local correlation feature in a compact basis of small molecules and construct an expressive low-data deep neural network (dNN) model to obtain machine-learned electron correlation energies on par with MP2 and CCSD levels of theory for more complex molecules and different datasets that are not represented in the training set. We show that our dNN-powered model is data efficient and makes highly transferable predictions across alkanes of various lengths, organic molecules with non-covalent and biomolecular interactions, as well as water clusters of different sizes and morphologies. In particular, by training 800 (H2O)8 clusters with the local correlation descriptors, accurate MP2/cc-pVTZ correlation energies up to (H2O)128 can be predicted with a small random error within chemical accuracy from exact values, while a majority of prediction deviations are attributed to an intrinsically systematic error. Our results reveal that an extremely compact local correlation feature set, which is poor for any direct post-Hartree-Fock calculations, has however a prominent advantage in reserving important electron correlation patterns for making accurate transferable predictions across distinct molecular compositions, bond types, and geometries.

IEnhance: a multi-scale spatial projection encoding network for enhancing chromatin interaction data resolution.

Although sequencing-based high-throughput chromatin interaction data are widely used to uncover genome-wide three-dimensional chromatin architecture, their sparseness and high signal-noise-ratio greatly restrict the precision of the obtained structural elements. To improve data quality, we here present iEnhance (chromatin interaction data resolution enhancement), a multi-scale spatial projection and encoding network, to predict high-resolution chromatin interaction matrices from low-resolution and noisy input data. Specifically, iEnhance projects the input data into matrix spaces to extract multi-scale global and local feature sets, then hierarchically fused these features by attention mechanism. After that, dense channel encoding and residual channel decoding are used to effectively infer robust chromatin interaction maps. iEnhance outperforms state-of-the-art Hi-C resolution enhancement tools in both visual and quantitative evaluation. Comprehensive analysis shows that unlike other tools, iEnhance can recover both short-range structural elements and long-range interaction patterns precisely. More importantly, iEnhance can be transferred to data enhancement of other tissues or cell lines of unknown resolution. Furthermore, iEnhance performs robustly in enhancement of diverse chromatin interaction data including those from single-cell Hi-C and Micro-C experiments.

Real-time detection of street tree crowns using mobile laser scanning based on pointwise classification

For targeted spraying in urban streets using mobile laser scanning (MLS), it is crucial to continuously detect tree crowns in real time from MLS data collected online. Crown detection based on a pointwise classification method, which predicts whether a point belongs to a crown using a binary classifier composed of a set of local features extracted from the neighbourhood of the point, is suitable for online data processing. However, the computational complexity of pointwise detection methods needs to be reduced to meet the real-time requirements of online targeted spraying. This paper proposes a real-time detection method for street tree crowns using MLS based on pointwise classification. A cubic neighbourhood is adopted from which a set of statistical features are extracted. A coarse-to-fine neighbourhood search method based on an MLS data grid storage structure is proposed to quickly query the neighbours. The classification performance and computational time of single features are evaluated. Then, the features are ranked using the technique for order preference by similarity to the ideal solution (TOPSIS). Finally, a subset of features that are easy to calculate and have high discrimination are selected and fused into a crown detector using a random forest algorithm. During online detection, the point cloud is sampled in terms of the vehicle movement and LiDAR measurement before being detected. An MLS point cloud collected from a 451 m urban street is used in the experiments. When the neighbourhood size is set to 0.3 m, the detection time per scanline is 24.84 ms on the unsampled test set with an F1 score of 0.9927. When the test set is sampled with a total sampling rate of 40, the detection time is reduced to 1% with an F1 score of 0.9906. The experimental results show that the proposed method can provide real-time and accurate detection of street tree crowns for online targeted spraying.

Independent mechanisms for processing local contour features and global shape.

The visual system can extract the global shape of an object from highly variable local contour features. We propose that there are separate systems for processing local and global shape. These systems are independent and process information differently. Global shape encoding accurately represents the form of low-frequency contour variations, whereas the local system encodes only summary statistics that describe typical features of high-frequency elements. In Experiments 1-4, we tested this hypothesis by obtaining same/different judgments for shapes that differed in local features, global features, or both. We found low sensitivity to changed local features that shared the same summary statistics, and no advantage in sensitivity for shapes that differed in both local and global features compared to shapes that differed only in global features. This sensitivity difference persisted when physical contour differences were equated and when shape feature sizes and exposure durations were increased. In Experiment 5, we compared sensitivity to sets of local contour features with matched or unmatched statistical properties. Sensitivity was higher for unmatched statistical properties than for properties sampled from the same statistical distribution. Experiment 6 directly tested our hypothesis of independent local and global systems using visual search. Search based on either local or global shape differences produced pop-out effects, but search for a target based on a conjunction of local and global differences required focal attention. These findings support the notion that separate mechanisms process local and global contour information and that the kinds of information these mechanisms encode are fundamentally different. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Local Binary Convolutional Neural Networks' Long Short-Term Memory Model for Human Embryos' Anomaly Detection

Accurate selection of embryos with the maximum implementation condition is a necessary step to increase the effectiveness of fertility treatment in in vitro fertilization (IVF). The deep learning algorithms presented high potential for monitoring and visualizing embryo features such as cell numbers and their morphological development in time series manner. Due to the ability of the computer vision and deep learning algorithms, this paper aimed to present a novel deep learning approach to distinguish simultaneous abnormality of embryos in time-lapse systems for detecting live and non-live births in IVF. The approach is composed of local binary convolutional neural network (LBCNN) and long short-term memory (LSTM). The LBCNN improved accuracy of classification by employing deep and local feature sets with lower number of learnable parameters in comparison with a standard convolutional layer. Moreover, LSTM network is employed to analyze temporal information of time-lapse embryos. The results indicate that the proposed approach achieves significant results in ROC analysis (0.98) in 5 days of monitoring compared to state-of-the-art models. In addition, the approach showed compatible results in early diagnosis of abnormality detection (72 hours) with 82.8% accuracy of classification compared to the pretrained well-known convolutional neural network (CNN) models and baseline CNN.

Assessment of shape-based features ability to predict the ascending aortic aneurysm growth.

The current guidelines for the ascending aortic aneurysm (AsAA) treatment recommend surgery mainly according to the maximum diameter assessment. This criterion has already proven to be often inefficient in identifying patients at high risk of aneurysm growth and rupture. In this study, we propose a method to compute a set of local shape features that, in addition to the maximum diameter D, are intended to improve the classification performances for the ascending aortic aneurysm growth risk assessment. Apart from D, these are the ratio DCR between D and the length of the ascending aorta centerline, the ratio EILR between the length of the external and the internal lines and the tortuosity T. 50 patients with two 3D acquisitions at least 6 months apart were segmented and the growth rate (GR) with the shape features related to the first exam computed. The correlation between them has been investigated. After, the dataset was divided into two classes according to the growth rate value. We used six different classifiers with input data exclusively from the first exam to predict the class to which each patient belonged. A first classification was performed using only D and a second with all the shape features together. The performances have been evaluated by computing accuracy, sensitivity, specificity, area under the receiver operating characteristic curve (AUROC) and positive (negative) likelihood ratio LHR+ (LHR-). A positive correlation was observed between growth rate and DCR (r = 0.511, p = 1.3e-4) and between GR and EILR (r = 0.472, p = 2.7e-4). Overall, the classifiers based on the four metrics outperformed the same ones based only on D. Among the diameter-based classifiers, k-nearest neighbours (KNN) reported the best accuracy (86%), sensitivity (55.6%), AUROC (0.74), LHR+ (7.62) and LHR- (0.48). Concerning the classifiers based on the four shape features, we obtained the best accuracy (94%), sensitivity (66.7%), specificity (100%), AUROC (0.94), LHR+ (+∞) and LHR- (0.33) with support vector machine (SVM). This demonstrates how automatic shape features detection combined with risk classification criteria could be crucial in planning the follow-up of patients with ascending aortic aneurysm and in predicting the possible dangerous progression of the disease.

Multifeature Fusion-Based Object Detection for Intelligent Transportation Systems

The detection of 3D objects with high precision from point cloud data has become a crucial research topic in intelligent transportation systems. By effectively modeling global and local features, it can be acquired the state-of-the-art detector for 3D object detection. Nevertheless, regarding the previous work on feature representations, volumetric generation or point learning methods have difficulty building the relationships between local features and global features. Thus, we propose a multi-feature fusion network (MFFNet) to improve detection precision for 3D point cloud data by combining the global features from 3D voxel convolutions with the local features from the point learning network. Our algorithm is an end-to-end detection framework that contains a voxel convolutional module, a local point feature module and a detection head. Significantly, MFFNet constructs the local point feature set with point learning and sampling and the global feature map through 3D voxel convolution from raw point clouds. The detection head can use the obtained fusion feature to predict the position and category of the examined 3D object, so the proposed method can obtain higher precision than existing approaches. An experimental evaluation on the KITTI 3D object detection dataset obtain 97% MAP (Mean Average Precision) and Waymo Open dataset obtain 80% MAP, which proves the efficiency of the developed feature fusion representation method for 3D objects, and it can achieve satisfactory location accuracy.

Discharge Summaries Based Sentiment Detection Using Multi-Head Attention and CNN-BiGRU

Automatic extraction of the patient’s health information from the unstructured data concerning the discharge summary remains challenging. Discharge summary related documents contain various aspects of the patient health condition to examine the quality of treatment and thereby help improve decision-making in the medical field. Using a sentiment dictionary and feature engineering, the researchers primarily mine semantic text features. However, choosing and designing features requires a lot of manpower. The proposed approach is an unsupervised deep learning model that learns a set of clusters embedded in the latent space. A composite model including Active Learning (AL), Convolutional Neural Network (CNN), BiGRU, and Multi-Attention, called ACBMA in this research, is designed to measure the quality of treatment based on discharge summaries text sentiment detection. CNN is utilized for extracting the set of local features of text vectors. Then BiGRU network was utilized to extract the text’s global features to solve the issues that a single CNN cannot obtain global semantic information and the traditional Recurrent Neural Network (RNN) gradient disappearance. Experiments prove that the ACBMA method can demonstrate the effectiveness of the suggested method, achieve comparable results to state-of-arts methods in sentiment detection, and outperform them with accurate benchmarks. Finally, several algorithm studies ultimately determined that the ACBMA method is more precise for discharge summaries sentiment analysis.

Does the brain's ventral visual pathway compute object shape?

A rich behavioral literature has shown that human object recognition is supported by a representation of shape that is tolerant to variations in an object's appearance. Such 'global' shape representations are achieved by describing objects via the spatial arrangement of their local features, or structure, rather than by the appearance of the features themselves. However, accumulating evidence suggests that the ventral visual pathway - the primary substrate underlying object recognition - may not represent global shape. Instead, ventral representations may be better described as a basis set of local image features. We suggest that this evidence forces a reevaluation of the role of the ventral pathway in object perception and posits a broader network for shape perception that encompasses contributions from the dorsal pathway.

A deep neural framework for real-time vehicular accident detection based on motion temporal templates

Vehicular accident prediction and detection has recently garnered curiosity and large amounts of attention in machine learning applications and related areas, due to its peculiar and fascinating application potentials in the development of Intelligent Transportation Systems (ITS) that play a pivotal role in the success of emerging smart cities. In this paper, we present a new vision-based framework for real-time vehicular accident prediction and detection, based on motion temporal templates and fuzzy time-slicing. The presented framework proceeds in a stepwise fashion, starting with automatically detecting moving objects (i.e., on-road vehicles or roadside pedestrians), followed by dynamically keep tracking of the detected moving objects based on temporal templates, clustering and supervised learning. Then, an extensive set of local features is extracted from the temporal templates of moving objects. Finally, an effective deep neural network (DNN) model is trained on the extracted features to detect abnormal vehicle behavioral patterns and thus predict an accident just before it occurs. The experiments on real-world vehicular accident videos demonstrate that the framework can yield mostly promising results by achieving a hit rate of 98.5% with a false alarm rate of 4.2% that compare very favorably to those from existing approaches, while still being able to deliver delay guarantees for realtime traffic monitoring and surveillance applications.

Automated quantitative assessment of amorphous calcifications: Towards improved malignancy risk stratification

BackgroundAmorphous calcifications noted on mammograms (i.e., small and indistinct calcifications that are difficult to characterize) are associated with high diagnostic uncertainty, often leading to biopsies. Yet, only 20% of biopsied amorphous calcifications are cancer. We present a quantitative approach for distinguishing between benign and actionable (high-risk and malignant) amorphous calcifications using a combination of local textures, global spatial relationships, and interpretable handcrafted expert features. MethodOur approach was trained and validated on a set of 168 2D full-field digital mammography exams (248 images) from 168 patients. Within these 248 images, we identified 276 image regions with segmented amorphous calcifications and a biopsy-confirmed diagnosis. A set of local (radiomic and region measurements) and global features (distribution and expert-defined) were extracted from each image. Local features were grouped using an unsupervised k-means clustering algorithm. All global features were concatenated with clustered local features and used to train a LightGBM classifier to distinguish benign from actionable cases. ResultsOn the held-out test set of 60 images, our approach achieved a sensitivity of 100%, specificity of 35%, and a positive predictive value of 38% when the decision threshold was set to 0.4. Given that all of the images in our test set resulted in a recommendation of a biopsy, the use of our algorithm would have identified 15 images (25%) that were benign, potentially reducing the number of breast biopsies. ConclusionsQuantitative analysis of full-field digital mammograms can extract subtle shape, texture, and distribution features that may help to distinguish between benign and actionable amorphous calcifications.

Frame-Based Locality Preservation Matching for Images Involving Large-Scale Transformations

Feature matching refers to the establishment of reliable correspondence between two sets of local features, which is an essential approach in remote sensing applications such as image registration and mosaicking. In this paper, a simple yet effective method, called frame-based locality preservation matching, is proposed for robust remote sensing image matching. We primarily focus on those images pairs that involve large-scale geometric transformations (e.g., extreme zoom). The key idea of our approach is to dig up the frame knowledge, such as the feature orientation and scale implied by common features like SIFT. The frame knowledge is free to obtain, and we find it to be of great significance in feature matching, especially for our focus -- large-scale geometric transformations. The proposed method can easily handle the geometric challenges and high outlier proportions, and significantly improves the performance compared to other state-of-the-art methods.

Three-Dimension Polarimetric Correlation Pattern Interpretation Tool and its Application

Polarimetric radar can acquire complete polarization information and is widely used in many applications. However, target orientation relative to the radar line of sight usually exhibits significant influences on the scattering mechanisms. Recently, such target scattering diversity has been successfully characterized and utilized with the polarimetric rotation domain interpretation techniques. In radar polarimetry, target scattering responses are affected by both polarization orientation angle and polarization ellipticity angle. In this vein, this work aims at exploring and utilizing the complete target scattering diversity by extending polarimetric rotation domain techniques to the polarization ellipticity angle dimension. The main idea is to develop a three-dimension polarimetric correlation pattern (3-D PCP) interpretation tool, which can visualize and exhibit targets’ polarimetric rotation domain properties in terms of both the polarimetric orientation and ellipticity angles. Then, a set of global, local, and mutual polarimetric features are proposed to characterize the responses of a 3-D PCP interpretation tool. Especially, the curvatures in differential geometry are first introduced to describe the properties of the 3-D surface. The performance of these new polarimetric features is investigated with spaceborne polarimetric synthetic aperture radar (PolSAR) data. Experimental results demonstrate the advantage of the proposed 3-D PCP features in enhancing the target clutter ratio (TCR), especially for the weak ship area. Based on this, a non-local superpixel-level contrast measure (NSLCM) method for ship detection is proposed. Pure sea samples can be determined adaptively for salient map construction. Comparison results demonstrate better detection performance for both the inshore dense ship area and weak ship area.

A CNN model embedded with local feature knowledge and its application to time-varying signal classification

A novel convolutional neural network is proposed for local prior feature embedding and imbalanced dataset modeling for multi-channel time-varying signal classification. This model consists of a single-channel signal feature parallel extraction unit, a multi-channel signal feature integration unit, a local feature embedding and feature similarity measurement unit, a full connection layer, and a Softmax classifier. An algorithm combining dynamic clustering and sliding window was used to select segments signals with typical local features in each pattern class, forming a typical local feature set. The one-dimensional CNNs were used to extract features from the single-channel signal in parallel, a comprehensive feature matrix of the multi-channel signal and the local feature matrix templates were produced. Using the method of external embedding, based on the sliding window and dynamic time warping (DTW) algorithm, the local feature similarities between the local feature template of each pattern class and the comprehensive feature sub-matrix of the input signal were measured, and the maximum values were selected to construct a local feature similarity vector in order. The information fusion was realized through a full connection layer. The proposed methodology can extract and represent both global and local signals features, strengthen the role of prior local feature in classification and improve the modeling properties of imbalanced datasets. A comprehensive learning algorithm is presented in this paper. The classification diagnosis of cardiovascular disease based on 12-lead ECG signals was used as a verification experiment. Results showed that the accuracy and generalization for the proposed technique were significantly improved.

A Feature Selection Method for Classifying Highly Similar Text Documents

In the era of big data, the importance of data classification is increasing. However, when it comes to classifying text documents, several obstacles degrade classification performance. These include multi-class documents, high levels of similarity between classes, class size imbalance, high dimensional representation space, and a low frequency of unique and discriminative features. To overcome these obstacles and improve classification performance, this paper proposes a novel feature selection method that effectively utilizes both unique and overlapping features. In general, feature selection methods have ignored unique features that occur only one class because of low frequency while it provides better discriminative-power. On the contrary, overlapping features, which are found in several classes with high frequency, have been also less preferred because of low discriminative-power. The proposed feature selection method attempts to use these two types of features as complementary with aims to improve overall classification performance for highly similar text documents. Extensive numerical analysis have been conducted for three benchmarking datasets with a support vector machine (SVM) classifier. The proposed method showed that not only the class with high similarity but also the general classification performance is superior to the conventional feature selection methods, such as the global feature set, local feature set, discriminative feature set, and information gain.

Pointwise classification of mobile laser scanning point clouds of urban scenes using raw data

Mobile laser scanning (MLS), which can quickly collect a high-resolution and high-precision point cloud of the surroundings of a vehicle, is an appealing technology for three-dimensional (3D) urban scene analysis. In this regard, the classification of MLS point clouds is a common and core task. We focus on pointwise classification, in which each individual point is categorized into a specific class by applying a binary classifier involving a set of local features derived from the neighborhoods of the point. To speed up the neighbor search and enhance feature distinctiveness for pointwise classification, we exploit the topological and semantic information in the raw data acquired by light detection and ranging (LiDAR) and recorded in scan order. First, a two-dimensional (2D) scan grid for data indexing is recovered, and the relative 3D coordinates with respect to the LiDAR position are calculated. Subsequently, a set of local features is extracted using an efficient neighbor search method with a low computational complexity independent of the number of points in a point cloud. These features are further merged to produce a variety of binary classifiers for specific classes via a GentleBoost supervised learning algorithm combining decision trees. The experimental results on the Paris-rue-Cassette database demonstrate that the proposed approach outperforms the state-of-the-art methods with a 10% improvement in the F1 score, whereas it uses simpler geometric features derived from a spherical neighborhood with a radius of 0.5 m.

Disparity-augmented trajectories for human activity recognition

Numerous methods for human activity recognition (HAR) have been proposed in the past two decades. Many of these methods are based on sparse representations, which describe the whole video content by a set of local features. Trajectories, as mid-level sparse features, are capable of describing the movements of interest points in two-dimensional (2D) space. However, 2D trajectories might be affected by viewpoint changes, potentially decreasing their accuracies. In this paper, we first propose and compare different 2D trajectory-based algorithms for human activity recognition. Then, we propose a new way of augmenting 2D trajectories with disparity information, without the calculation of the 3D reconstruction. Our obtained HAR results have shown a 2.76% improvement when using disparity-augmented trajectories, compared to using classical 2D trajectory information only. Furthermore, we have also tested our method on the challenging Hollywood 3D dataset, on which we have obtained competitive results, at a much faster speed.