Human body motion capture from multi-image video sequences

In this paper we present a method to extract 3-D information of the shape and movement of the human body using video sequences acquired with three CCD cameras. This work is part of a project aimed at developing a highly automated system to model most realistically human bodies from video sequences. Our image acquisition system is currently composed of three synchronized CCD cameras and a frame grabber which acquires a sequence of triplet images. From the video sequences, we extract two kinds of 3-D information: a three dimensional surface measurement of the visible body parts for each triplet and 3-D trajectories of points on the body. Our approach to surface measurement is based on multi-image matching, using the adaptive least squares method. A semi automated matching process determines a dense set of corresponding points in the triplets, starting from few manually selected seed points. The tracking process is also based on least squares matching techniques, thus the name LSMTA (Least Squares Matching Tracking Algorithm). The spatial correspondences between the three images of the different views and the temporal correspondences between subsequent frames are determined with a least squares matching algorithm. The advantage of this tracking process is twofold: firstly, it can track natural points, without using markers; secondly, it can also track entire surface parts on the human body. In the last case, the tracking process is applied to all the points matched in the region of interest. The result can be seen as a vector field of trajectories (position, velocity and acceleration) which can be checked with thresholds and neighborhood-based filters. Key points on the body can be defined in the vector field of trajectories and tracked through the sequence leading to a result comparable to the conventional motion capture systems.

Ground target detection is very important in precise infrared imaging guidance. To address this problem, an accurate tracking algorithm of the key points, i.e., vertex of buildings is proposed. First, the feature points are extracted by Kanade-Lucas-Tomasi (KLT) algorithm, and the template of feature points is updated constantly in the tracking process according to the offset. Then based on the extracted feature points, the key point can be positioned using the geometric relation between the feature points and the key point. Third, the algorithm tracks the feature points and uses the geometric relation to track the key point in the next frame. The experimental results demonstrate the effectiveness of the proposed algorithm in tracking the key point of buildings in front-lower infrared image sequences for long time precise guidance.

Abstract. In this article the new method for iris image features extraction based on phase congruency is proposed. Iris image key points are calculated using the convolutions with Hermite transform functions. At each key point the feature vector characterizing this key point is obtained based on the phase congruency method. Iris key point descriptor contains phase congruency values at points located on concentric circles around the key point. To compare the key points, Euclidean metric between the key points descriptors is calculated. The distance between the iris images is equal to the number of matched iris key points. The proposed method was tested using the images from CASIA−IrisV4−Interval database and the value of EER = 0.226% was obtained.

Objective To explore the effects of assessment system based on key points in clinical skills training among nursing interns. Methods Nursing interns of the Affiliated Cancer Hospital of Zhengzhou University were selected as study subjects from January to December 2016 by the method of convenience sampling. They were divided into experimental group (n=50, practice started from January 2016) and control group (n=50, practice started from June 2016) according to the order of internship. Students in control group adopted the conventional skill training and assessment, while students in experimental group received the assessment system based on key points. Critical thinking skills of nursing interns of both groups were compared with the critical thinking disposition inventory (CTDI) . Results After training, the total score of CTDI of students (299.82±26.95) in experimental group was higher than that (283.41±24.24) in control group with a significant difference (t=3.201, P=0.002) . The scores of 7 dimensions in CTDI were (41.18±4.12) , (40.27±3.16) , (46.10±2.00) , (40.11±5.31) , (42.97±5.05) , (43.96±3.88) , (45.23±3.08) significantly higher than those in control group[ (39.62±3.09) , (38.62±4.15) , (42.04±2.95) , (37.92±3.65) , (39.50±3.57) , (41.82±5.17) , (43.51±2.64) ] (t=2.142, 2.237, 8.074, 2.402, 3.967, 2.341, 2.981; P<0.05) . Conclusions The assessment system based on key points can effectively improve the critical thinking skills of nursing interns. It is worth to be used widely. Key words: Nursing care; Intern; Key point; Clinical skill; Critical thinking

With the rise of computer vision, it is becoming more and more important to accurately recognize and evaluate human actions. However, the complexity, intraclass differences, and viewing angle changes of human actions significantly impact the accuracy of identification human actions. This paper proposed an action recognition and evaluation method based on skeleton information extraction. Briefly, We use Lightweight OpenPose to extract the key points of human skeleton and perform processing work, including video data cutting, deleting some key points, supplementing missing key points, filtering processing, feature extraction, etc. Through an in-depth exploration of related theoretical technologies, we proposed a model for recognition and evaluation of human table tennis actions with an ordinary camera. The support vector machine algorithm (SVM) classification model is used to identify table tennis actions in real-time. Then the dynamic time regularization (DTW) algorithm calculates the similarity of each human skeleton key point in the action sequence. The low-scoring bone key points are marked to evaluate the human table tennis action in real time. The results show that a recognition rate of more than 95% is achieved in the test set, which proves the method's effectiveness. In addition, we compared the results with previous work using inertial sensors for action recognition, which shows our method can preserve the same accuracy with a much lower cost of implementation.

BackgroundDrosophila melanogaster is an important organism used in many fields of biological research such as genetics and developmental biology. Drosophila wings have been widely used to study the genetics of development, morphometrics and evolution. Therefore there is much interest in quantifying wing structures of Drosophila. Advancement in technology has increased the ease in which images of Drosophila can be acquired. However such studies have been limited by the slow and tedious process of acquiring phenotypic data.ResultsWe have developed a system that automatically detects and measures key points and vein segments on a Drosophila wing. Key points are detected by performing image transformations and template matching on Drosophila wing images while vein segments are detected using an Active Contour algorithm.The accuracy of our key point detection was compared against key point annotations of users. We also performed key point detection using different training data sets of Drosophila wing images. We compared our software with an existing automated image analysis system for Drosophila wings and showed that our system performs better than the state of the art. Vein segments were manually measured and compared against the measurements obtained from our system.ConclusionOur system was able to detect specific key points and vein segments from Drosophila wing images with high accuracy.

Biharmonic deformation transfer with automatic key point selection

Abstract. Remote sensing techniques are an important tool in fluvial transport monitoring, since they allow for effective evaluation of the volume of transported material. Nevertheless, there is no methodology for automatic calculation of movement parameters of individual rocks. These parameters can be determined by point cloud registration. Hence, the goal of this study is to develop a robust algorithm for terrestrial laser scanning point cloud registration. The registration is based on Iterative Closest Point algorithm, which requires well established initial parameters of transformation. Thus, we propose to calculate the initial parameters based on key points representing the maximum of Gaussian curvature. For each key point the set of geometrical features is calculated. The key points are then matched between two point clouds as a nearest neighbor in feature domain. Different combinations of neighborhood sizes, feature subsets, metrics and number of nearest neighbors were tested to obtain the highest ratio between properly and improperly matched key points. Finally, RANSAC algorithm was used to calculate the initial transformation parameters between the point clouds and the ICP algorithm was used for calculation of final transformation parameters. The investigations carried out on sample point clouds representing rocks enabled the adjustment of parameters of the algorithm and showed that the Gaussian curvature can be used as a 3-dimentional key point detector for such objects. The proposed algorithm enabled to register point clouds with the mean distance between point clouds equal to 3 mm.

In this paper we introduce a graph grammar based method to fuse the low level features and apply them to object detecting and tracking. In our algorithm, the graph grammar rules are used to detect the object in the beginning of the video sequence and then dynamically adjust the tracking procedure. Our tracking algorithm consists of two phases: key points tracking and tracking by graph grammar rules. The key points are computed by using salient level set components. All key points, as well as the colors and the tangent directions, are fed to a Kalman filter for object tracking. Then the graph grammar rules are used to dynamically examine and adjust the tracking procedure to make it robust. The effectiveness of the algorithm has been demonstrated by experiments.

We address a core problem of computer vision: Detection and description of 2D feature points for image matching. For a long time, hand-crafted designs, like the seminal SIFT algorithm, were unsurpassed in accuracy and efficiency. Recently, learned feature detectors emerged that implement detection and description using neural networks. Training these networks usually resorts to optimizing low-level matching scores, often pre-defining sets of image patches which should or should not match, or which should or should not contain key points. Unfortunately, increased accuracy for these low-level matching scores does not necessarily translate to better performance in high-level vision tasks. We propose a new training methodology which embeds the feature detector in a complete vision pipeline, and where the learnable parameters are trained in an end-to-end fashion. We overcome the discrete nature of key point selection and descriptor matching using principles from reinforcement learning. As an example, we address the task of relative pose estimation between a pair of images. We demonstrate that the accuracy of a state-of-the-art learning-based feature detector can be increased when trained for the task it is supposed to solve at test time. Our training methodology poses little restrictions on the task to learn, and works for any architecture which predicts key point heat maps, and descriptors for key point locations.

Key-point based automated diagnosis for alveolar dehiscence in mandibular incisors using convolutional neural network

Lightning protection design for 10kV composite distribution tower differs from transmission towers and the traditional iron tower due to the insulation of composite material and complexity of lightning protection design for 10kV distribution lines. Related research is still in blank. Key points of lightning protection design are discussed based on the simulation study on the overvoltage characteristic of direct lightning and induction lightning of 10kV composite distribution tower with or without the grounding wire. The result shows that composite tower without grounding wire is suitable for regions where induction lightning is the main risk and the key point is to transfer the flashover path in order to avoid the arc burning the composite tower; Composite tower with grounding wire is suitable for regions where direct lightning is the main risk and the key points are to improve the insulation strength between phase and the ground in order to enhance the induction lightning withstanding level and to improve the insulation strength between phases to avoid double phase flashover before the release of the lightning current. According to the key points, structural parameter optimization is carried out.

In this paper we introduce a new dataset and pose invariant sampling method and describe the ensemble methods used for recognizing faces in 3D scenes, captured using commodity depth sensors. We use the 3D SIFT key point detector to take advantage of the similarities between faces, which leads to a set of points of interest based on the curvature of the face. For all key points, features are extracted using a 3D feature descriptor. Then, a variable-sized amount of features are generated per each 3D face image. The first ensemble method we constructed uses a K-nearest neighbors classifier to classify each key point-sampled feature vector as belonging to one of the subjects recorded in our dataset. All votes over all key points are combined. In the second ensemble technique, the key points are clustered with K-means, using the feature vectors and approximated sampling positions relative to the face. This leads to a set of experts that specialize for a specific region. Then a K-nearest neighbors classifier is trained on the examples falling in each expert's specialized region. Finally, for a new 3D face image, votes from all experts are combined in a sum ensemble technique to categorize the 3D face. We also introduce 6 new real world datasets with different variances: 3 types of 3D rotations, distance to sensor, expressions, and an all-in-one dataset. The results show very high cross validation accuracies for the same type of variance. In addition, 36 variance specific pair-tests in which the system is trained on one dataset and tested on a completely different dataset also show encouraging results.

We present the system description for our submission towards the Key Point Analysis Shared Task at ArgMining 2021. Track 1 of the shared task requires participants to develop methods to predict the match score between each pair of arguments and key points, provided they belong to the same topic under the same stance. We leveraged existing state of the art pre-trained language models along with incorporating additional data and features extracted from the inputs (topics, key points, and arguments) to improve performance. We were able to achieve mAP strict and mAP relaxed score of 0.872 and 0.966 respectively in the evaluation phase, securing 5th place on the leaderboard. In the post evaluation phase, we achieved a mAP strict and mAP relaxed score of 0.921 and 0.982 respectively.

Geodesic-like features for point matching