Automatic analysis of complex athlete techniques in broadcast taekwondo video

Athlete detection and action recognition in sports video is a very challenging task due to the dynamic and cluttered background. Several attempts for automatic analysis focus on athletes in many sports videos have been made. However, taekwondo video analysis remains an unstudied field. In light of this, a novel framework for automatic techniques analysis in broadcast taekwondo video is proposed in this paper. For an input video, in the first stage, athlete tracking and body segmentation are done through a modified Structure Preserving Object Tracker. In the second stage, the de-noised frames which completely contain the body of analyzed athlete from video sequence, are trained by a deep learning network PCANet to predict the athlete action of each single frame. As one technique is composed of many consecutive actions and each action corresponds a video frame, focusing on video sequences to achieve techniques analysis makes sense. In the last stage, linear SVM is used with the predicted action frames to get a techniques classifier. To evaluate the performance of the proposed framework, extensive experiments on real broadcast taekwondo video dataset are provided. The results show that the proposed method achieves state-of-the-art results for complex techniques analysis in taekwondo video.

     

3D human action analysis and recognition through GLAC descriptor on 2D motion and static posture images

In this paper, we present an approach for identification of actions within depth action videos. First, we process the video to get motion history images (MHIs) and static history images (SHIs) corresponding to an action video based on the use of 3D Motion Trail Model (3DMTM). We then characterize the action video by extracting the Gradient Local Auto-Correlations (GLAC) features from the SHIs and the MHIs. The two sets of features i.e., GLAC features from MHIs and GLAC features from SHIs are concatenated to obtain a representation vector for action. Finally, we perform the classification on all the action samples by using the l2-regularized Collaborative Representation Classifier (l2-CRC) to recognize different human actions in an effective way. We perform evaluation of the proposed method on three action datasets, MSR-Action3D, DHA and UTD-MHAD. Through experimental results, we observe that the proposed method performs superior to other approaches.

     

动作切分和流形度量学习的视频动作识别

目的 为了提高视频中动作识别的准确度,提出基于动作切分和流形度量学习的视频动作识别算法。方法 首先利用基于人物肢体伸展程度分析的动作切分方法对视频中的动作进行切分,将动作识别的对象具体化;然后从动作片段中提取归一化之后的全局时域特征和空域特征、光流特征、帧内的局部旋度特征和散度特征,构造一种7×7的协方差矩阵描述子对提取出的多种特征进行融合;最后结合流形度量学习方法有监督式地寻找更优的距离度量算法提高动作的识别分类效果。结果 对Weizmann公共视频集的切分实验统计结果表明本文提出的视频切分方法具有很好的切分能力,能够作好动作识别前的预处理;在Weizmann公共视频数据集上进行了流形度量学习前后的识别效果对比,结果表明利用流形度量学习方法对动作识别效果提升2.8%;在Weizmann和KTH两个公共视频数据集上的平均识别率分别为95.6%和92.3%,与现有方法的比较表明,本文提出的动作识别方法有更好的识别效果。结论 多次实验结果表明本文算法在预处理过程中动作切分效果理想,描述动作所构造协方差矩阵对动作的表达有良好的多特征融合能力,而且光流信息和旋度、散度信息的加入使得人体各部位的运动方向信息具有了更多细节的描述,有效提高了协方差矩阵的描述能力,结合流形度量学习方法对动作识别的准确性有明显提高。     

基于播音员识别的新闻视频故事分割方法

新闻视频的语义单元分割是基于内容的新闻视频检索和情报挖掘的重要步骤,受到众多研究者的关注。提出了一种基于播音员识别的新闻视频故事单分割的新方法,首先从新闻节目中提取各播音员的声学感知特征的作为其声纹,训练出其相应的混合高斯模型(GMM),并采用KL差异法从视频镜头中探测出各播音员和非播音员音频镜头,最后结合视频字幕帧事件和新闻节目特殊的结构知识对新闻节目进行故事单元分割。在2个多小时的CCTV和CNN新闻视频实验中获得96.02%查准率和92.58%的查全率。     

Twin Gaussian Processes for Structured Prediction

We describe twin Gaussian processes (TGP), a generic structured prediction method that uses Gaussian process (GP) priors on both covariates and responses, both multivariate, and estimates outputs by minimizing the Kullback-Leibler divergence between two GP modeled as normal distributions over finite index sets of training and testing examples, emphasizing the goal that similar inputs should produce similar percepts and this should hold, on average, between their marginal distributions. TGP captures not only the interdependencies between covariates, as in a typical GP, but also those between responses, so correlations among both inputs and outputs are accounted for. TGP is exemplified, with promising results, for the reconstruction of 3d human poses from monocular and multicamera video sequences in the recently introduced HumanEva benchmark, where we achieve 5 cm error on average per 3d marker for models trained jointly, using data from multiple people and multiple activities. The method is fast and automatic: it requires no hand-crafting of the initial pose, camera calibration parameters, or the availability of a 3d body model associated with human subjects used for training or testing.     

Unsupervised Learning of Human Action Categories Using Spatial-Temporal Words

We present a novel unsupervised learning method for human action categories. A video sequence is represented as a collection of spatial-temporal words by extracting space-time interest points. The algorithm automatically learns the probability distributions of the spatial-temporal words and the intermediate topics corresponding to human action categories. This is achieved by using latent topic models such as the probabilistic Latent Semantic Analysis (pLSA) model and Latent Dirichlet Allocation (LDA). Our approach can handle noisy feature points arisen from dynamic background and moving cameras due to the application of the probabilistic models. Given a novel video sequence, the algorithm can categorize and localize the human action(s) contained in the video. We test our algorithm on three challenging datasets: the KTH human motion dataset, the Weizmann human action dataset, and a recent dataset of figure skating actions. Our results reflect the promise of such a simple approach. In addition, our algorithm can recognize and localize multiple actions in long and complex video sequences containing multiple motions.     

误差准则的信息论解释

Error criteria (or error cost functions) play significant roles in statistical estimation problems. In this paper, we study error criteria from the viewpoint of information theory. The relationships between error criteria and error's entropy criterion are investigated. It is shown that an error criterion is equivalent to the error's entropy criterion plus a Kullback-Leibler information divergence (KL-divergence). Based on this result, two important properties of the error criteria are proved. Particularly, the optimum error criterion can be interpreted via the meanings of entropy and KL-divergence. Furthermore, a novel approach is proposed for the choice of p-power error criteria, in which a KL-divergence based cost is minimized. The proposed method is verified by Monte Carlo simulation experiments.     

Human action recognition using three orthogonal planes with unsupervised deep convolutional neural network

Deep learning models have attained great success for an extensive range of computer vision applications including image and video classification. However, the complex architecture of the most recently developed networks imposes certain memory and computational resource limitations, especially for human action recognition applications. Unsupervised deep convolutional neural networks such as PCANet can alleviate these limitations and hence significantly reduce the computational complexity of the whole recognition system. In this work, instead of using 3D convolutional neural network architecture to learn temporal features of video actions, the unsupervised convolutional PCANet model is extended into (PCANet-TOP) which effectively learn spatiotemporal features from Three Orthogonal Planes (TOP). For each video sequence, spatial frames (XY) and temporal planes (XT and YT) are utilized to train three different PCANet models. Then, the learned features are fused after reducing their dimensionality using whitening PCA to obtain spatiotemporal feature representation of the action video. Finally, Support Vector Machine (SVM) classifier is applied for action classification process. The proposed method is evaluated on four benchmarks and well-known datasets, namely, Weizmann, KTH, UCF Sports, and YouTube action datasets. The recognition results show that the proposed PCANet-TOP provides discriminative and complementary features using three orthogonal planes and able to achieve promising and comparable results with state-of-the-art methods.

     

Human action recognition using an ensemble of body‐part detectors

This paper describes an approach to human action recognition based on a probabilistic optimization model of body parts using hidden Markov model (HMM). Our method is able to distinguish between similar actions by only considering the body parts having major contribution to the actions, for example, legs for walking, jogging and running; arms for boxing, waving and clapping. We apply HMMs to model the stochastic movement of the body parts for action recognition. The HMM construction uses an ensemble of body‐part detectors, followed by grouping of part detections, to perform human identification. Three example‐based body‐part detectors are trained to detect three components of the human body: the head, legs and arms. These detectors cope with viewpoint changes and self‐occlusions through the use of ten sub‐classifiers that detect body parts over a specific range of viewpoints. Each sub‐classifier is a support vector machine trained on features selected for the discriminative power for each particular part/viewpoint combination. Grouping of these detections is performed using a simple geometric constraint model that yields a viewpoint‐invariant human detector. We test our approach on three publicly available action datasets: the KTH dataset, Weizmann dataset and HumanEva dataset. Our results illustrate that with a simple and compact representation we can achieve robust recognition of human actions comparable to the most complex, state‐of‐the‐art methods.     

自注意力机制的视频摘要模型

针对如何高效地识别出视频中具有代表性的内容问题,提出了一种对不同的视频帧赋予不同重要性的视频摘要算法.首先使用长短期记忆网络来建模视频序列的时序关系,然后利用自注意力机制建模视频中不同帧的重要性程度并提取全局特征,最后通过每一帧回归得到的重要性得分进行采样,并使用强化学习策略优化模型参数.其中,强化学习的动作定义为每一帧选或者不选,状态定义为当前这个视频的选择情况,反馈信号使用多样性和代表性代价.在2个公开数据集SumMe和TVSum中进行视频摘要实验,并使用F-度量来衡量这2个数据集上不同视频摘要算法的准确度,实验结果表明,提出的视频摘要算法结果要优于其他算法.     

基于PoseC3D的网球动作识别及评价方法

为了准确地识别及评价网球动作,将计算机视觉与网球运动相关知识相结合,提出了一种基于PoseC3D的网球动作识别及评价方法。首先,使用基于ResNet-50姿态估计模型对网球运动视频进行人体目标检测并提取骨骼关键点;然后,使用在专业网球场采集的视频数据集进行PoseC3D模型训练,使模型能够对网球的子动作进行分类;之后,使用动态时间规整算法对分类的动作进行评价;最后,基于采集的视频数据集进行了大量实验。结果表明,提出的基于PoseC3D的网球动作识别方法对6类网球子动作的分类Top1准确率可以达到90.8%。相较于基于图卷积网络的方法,比如AGCN和ST-GCN,具有更强的泛化能力;提出的基于动态时间规整的评分算法能够在动作分类后实时、准确地给出相应动作的评价分数,从而减少了网球教师的工作强度,有效地提升了网球教学质量。     

Human action recognition based on skeleton splitting

Human action recognition, defined as the understanding of the human basic actions from video streams, has a long history in the area of computer vision and pattern recognition because it can be used for various applications. We propose a novel human action recognition methodology by extracting the human skeletal features and separating them into several human body parts such as face, torso, and limbs to efficiently visualize and analyze the motion of human body parts.Our proposed human action recognition system consists of two steps: (i) automatic skeletal feature extraction and splitting by measuring the similarity between neighbor pixels in the space of diffusion tensor fields, and (ii) human action recognition by using multiple kernel based Support Vector Machine. Experimental results on a set of test database show that our proposed method is very efficient and effective to recognize the actions using few parameters.     

A component-based video content representation for action recognition

This paper investigates the challenging problem of action recognition in videos and proposes a new component-based approach for video content representation. Although satisfactory performance for action recognition has already been obtained for certain scenarios, many of the existing solutions require fully-annotated video datasets in which region of the activity in each frame is specified by a bounding box. Another group of methods require auxiliary techniques to extract human-related areas in the video frames before being able to accurately recognize actions. In this paper, a Weakly-Supervised Learning (WSL) framework is introduced that eliminates the need for per-frame annotations and learns video representations that improve recognition accuracy and also highlights the activity related regions within each frame. To this end, two new representation ideas are proposed, one focus on representing the main components of an action, i.e. actionness regions, and the other focus on encoding the background context to represent general and holistic cues. A three-stream CNN is developed, which takes the two proposed representations and combines them with a motion-encoding stream. Temporal cues in each of the three different streams are modeled through LSTM, and finally fully-connected neural network layers are used to fuse various streams and produce the final video representation. Experimental results on four challenging datasets, demonstrate that the proposed Component-based Multi-stream CNN model (CM-CNN), trained on a WSL setting, outperforms the state-of-the-art in action recognition, even the fully-supervised approaches.     

Triplet attention multiple spacetime-semantic graph convolutional network for skeleton-based action recognition

Skeleton-based action recognition has recently attracted widespread attention in the field of computer vision. Previous studies on skeleton-based action recognition are susceptible to interferences from redundant video frames in judging complex actions but ignore the fact that the spatial-temporal features of different actions are extremely different. To solve these problems, we propose a triplet attention multiple spacetime-semantic graph convolutional network for skeleton-based action recognition (AM-GCN), which can not only capture the multiple spacetime-semantic feature from the video images to avoid limited information diversity from single-layer feature representation but can also improve the generalization ability of the network. We also present the triplet attention mechanism to apply an attention mechanism to different key points, key channels, and key frames of the actions, improving the accuracy and interpretability of the judgement of complex actions. In addition, different kinds of spacetime-semantic feature information are combined through the proposed fusion decision for comprehensive prediction in order to improve the robustness of the algorithm. We validate AM-GCN with two standard datasets, NTU-RGBD and Kinetics, and compare it with other mainstream models. The results show that the proposed model achieves tremendous improvement.

     

Symbolic quality control for multimedia applications

We present a fine grain quality control method for multimedia applications. The method takes as input an application software composed of actions. The execution times of actions are unknown increasing functions of quality level parameters. The method allows the construction of a Controller which computes adequate action schedules and corresponding quality levels, so as to meet QoS requirements for a given platform. These include requirements for safety (action deadlines are met) as well optimality (maximization and smoothness of quality levels). The Controller consists of a Quality Manager and a Scheduler. For each action, the Controller uses a quality management policy for choosing a schedule and quality levels meeting the QoS requirements. The schedule is selected amongst a set of optimal schedules computed by the Scheduler. We extend and improve results of previous papers providing a solid theoretical basis for designing and implementing the Controller. We propose a symbolic quality management method using speed diagrams, a representation of the controlled system’s dynamics. Instead of numerically computing a quality level for each action, the Quality Manager changes action quality levels based on the knowledge of constraints characterizing control relaxation regions. These are sets of states in which quality management for a given number of computation steps can be relaxed without degrading quality. We study techniques for efficient computation of optimal schedules. We present experimental results including the implementation of the method and benchmarks for an MPEG4 video encoder. The benchmarks show drastic performance improvement for controlled quality with respect to constant quality. They also show that symbolic quality management allows significant reduction of the overhead with respect to numeric quality management. Finally, using optimal schedules can lead to considerable performance gains.     

Human action recognition using boosted EigenActions

This paper proposes a boosting EigenActions algorithm for human action recognition. A spatio-temporal Information Saliency Map (ISM) is calculated from a video sequence by estimating pixel density function. A continuous human action is segmented into a set of primitive periodic motion cycles from information saliency curve. Each cycle of motion is represented by a Salient Action Unit (SAU), which is used to determine the EigenAction using principle component analysis. A human action classifier is developed using multi-class Adaboost algorithm with Bayesian hypothesis as the weak classifier. Given a human action video sequence, the proposed method effectively locates the SAUs in the video, and recognizes the human actions by categorizing the SAUs. Two publicly available human action databases, namely KTH and Weizmann, are selected for evaluation. The average recognition accuracy are 81.5% and 98.3% for KTH and Weizmann databases, respectively. Comparative results with two recent methods and robustness test results are also reported.     

基于局部时空特征方向加权的人体行为识别

目的 对人体行为的描述是行为识别中的关键问题,为了能够充分利用训练数据从而保证特征对行为的高描述性,提出了基于局部时空特征方向加权的人体行为识别方法。方法 首先,将局部时空特征的亮度梯度特征分解为3个方向(X、Y、Z)分别来描述行为, 通过直接构造视觉词汇表分别得到不同行为3方向特征描述子集合的标准视觉词汇码本,并利用训练视频得到每个行为的标准3方向词汇分布;进而,根据不同行为3方向特征描述子集合的标准视觉词汇码本,分别计算测试视频相应的3方向的词汇分布,并利用与各行为标准3方向词汇分布的加权相似性度量进行行为识别;结果 在Weizmann数据库和KTH数据库中进行实验,Weizmann数据库中的平均识别率高达96.04%,KTH数据库中的平均识别率也高达96.93%。结论 与其他行为识别方法相比可以明显提高行为平均识别率。     

Difference intensity distance group pattern for recognizing actions in video using Support Vector Machines

Recognition of human actions is a very important, task in many applications such as Human Computer Interaction, Content based video retrieval and indexing, Intelligent video surveillance, Gesture Recognition, Robot learning and control, etc. An efficient action recognition system using Difference Intensity Distance Group Pattern (DIDGP) method and recognition using Support Vector Machines (SVM) classifier is presented. Initially, Region of Interest (ROI) is extracted from the difference frame, where it represents the motion information. The extracted ROI is divided into two blocks B1 and B2. The proposed DIDGP feature is applied on the maximum intensity block of the ROI to discriminate the each action from video sequences. The feature vectors obtained from the DIDGP are recognized using SVM with polynomial and RBF kernel. The proposed work has been evaluated on KTH action dataset which consists of actions like walking, running, jogging, hand waving, clapping and boxing. The proposed method has been experimentally tested on KTH dataset and an overall accuracy of 94.67% for RBF kernel.     

Prediction of Manipulation Actions

By looking at a person’s hands, one can often tell what the person is going to do next, how his/her hands are moving and where they will be, because an actor’s intentions shape his/her movement kinematics during action execution. Similarly, active systems with real-time constraints must not simply rely on passive video-segment classification, but they have to continuously update their estimates and predict future actions. In this paper, we study the prediction of dexterous actions. We recorded videos of subjects performing different manipulation actions on the same object, such as “squeezing”, “flipping”, “washing”, “wiping” and “scratching” with a sponge. In psychophysical experiments, we evaluated human observers’ skills in predicting actions from video sequences of different length, depicting the hand movement in the preparation and execution of actions before and after contact with the object. We then developed a recurrent neural network based method for action prediction using as input image patches around the hand. We also used the same formalism to predict the forces on the finger tips using for training synchronized video and force data streams. Evaluations on two new datasets show that our system closely matches human performance in the recognition task, and demonstrate the ability of our algorithms to predict in real time what and how a dexterous action is performed.     

On robustness of radial basis function network with input perturbation

In this article, we have proposed a methodology for making a radial basis function network (RBFN) robust with respect to additive and multiplicative input noises. This is achieved by properly selecting the centers and widths for the radial basis function (RBF) units of the hidden layer. For this purpose, firstly, a set of self-organizing map (SOM) networks are trained for center selection. For training a SOM network, random Gaussian noise is injected in the samples of each class of the data set. The number of SOM networks is same as the number of classes present in the data set, and each of the SOM networks is trained separately by the samples belonging to a particular class. The weight vector associated with a unit in the output layer of a particular SOM network corresponding to a class is used as the center of a RBF unit for that class. To determine the widths of the RBF units, p-nearest neighbor algorithm is used class-wise. Proper selection of centers and widths makes the RBFN robust with respect to input perturbation and outliers present in the data set. The weights between the hidden and output layers of RBFN are obtained by pseudo inverse method. To test the robustness of the proposed method in additive and multiplicative noise scenarios, ten standard data sets have been used for classification. Proposed method has been compared with three existing methods, where the centers have been generated in three ways: randomly, using k-means algorithm, and based on SOM network. Simulation results show the superiority of the proposed method compared to those methods. Wilcoxon signed-rank test also shows that the proposed method is statistically better than those methods.