共查询到20条相似文献,搜索用时 15 毫秒
1.
针对当前人体动作识别算法不能实现准确快速识别的问题,提出了一种基于特征蒸馏的实时人体动作识别算法。首先,针对C3D算法准确率较低的问题,引入通道注意力机制,改善网络的性能,得到SEC3D后作为教师网络;然后,利用SEC3D指导学生网络的学习,将教师网络的知识通过蒸馏算法迁移到学生网络中。在动作视频数据集UCF101和HMDB51上进行实验,学生网络在比教师网络模型压缩了23.7倍、特征维度压缩了25%的情况下,检测速度达到358.7f/s,满足实时性的要求。实验结果表明,该算法在远高于原始C3D算法检测速度的同时,在尽量减少精度损失下,减少模型参数和计算量,构建了一种轻量型的精度和速度共存的实时人体动作识别模型。 相似文献
2.
3.
The rapid emergence of pesticide resistance has given rise to a demand for herbicides with new mode of action (MoA). In the agrochemical sector, with the availability of experimental high throughput screening (HTS) data, it is now possible to utilize in silico target prediction methods in the early discovery phase to suggest the MoA of a compound via data mining of bioactivity data. While having been established in the pharmaceutical context, in the agrochemical area this approach poses rather different challenges, as we have found in this work, partially due to different chemistry, but even more so due to different (usually smaller) amounts of data, and different ways of conducting HTS. With the aim to apply computational methods for facilitating herbicide target identification, 48,000 bioactivity data against 16 herbicide targets were processed to train Laplacian modified Naïve Bayesian (NB) classification models. The herbicide target prediction model (“HerbiMod”) is an ensemble of 16 binary classification models which are evaluated by internal, external and prospective validation sets. In addition to the experimental inactives, 10,000 random agrochemical inactives were included in the training process, which showed to improve the overall balanced accuracy of our models up to 40%. For all the models, performance in terms of balanced accuracy of ≥ 80% was achieved in five-fold cross validation. Ranking target predictions was addressed by means of z-scores which improved predictivity over using raw scores alone. An external testset of 247 compounds from ChEMBL and a prospective testset of 394 compounds from BASF SE tested against five well studied herbicide targets (ACC, ALS, HPPD, PDS and PROTOX) were used for further validation. Only 4% of the compounds in the external testset lied in the applicability domain and extrapolation (and correct prediction) was hence impossible, which on one hand was surprising, and on the other hand illustrated the utilization of using applicability domains in the first place. However, performance better than 60% in balanced accuracy was achieved on the prospective testset, where all the compounds fell within the applicability domain, and which hence underlines the possibility of using target prediction also in the area of agrochemicals. 相似文献
4.
蛇怪蜥蜴踏水奔跑过程中,脚掌会在气穴封闭前移出水面,并不断调整姿态适应变化的流场.为分析踏水机理,增加了对出水过程的研究,建立了脚掌进出水流体动力学模型,运用RNG(renormalization group)k-ε湍流方程,结合气穴扩张收缩变化,获得了最优上提时间.进一步针对脚掌姿态调整行为,解析了3维驱动力随入水角度的变化规律,推导了托举力与速度的数学函数,并通过实验初步验证了数值计算模型的正确性.以蛇怪蜥蜴为仿生对象,将脚掌往复踏水转变为叶片旋转击水,设计了水面矢量推进器,增加了驱动力输出维数,建立流体动力学模型,分析了叶片数量对力学性能的影响.进行推进器应用实验,实现了水面平台在10.6°仰角下的滑水航行. 相似文献
5.
6.
View Invariance for Human Action Recognition 总被引:4,自引:0,他引:4
This paper presents an approach for viewpoint invariant human action recognition, an area that has received scant attention
so far, relative to the overall body of work in human action recognition. It has been established previously that there exist
no invariants for 3D to 2D projection. However, there exist a wealth of techniques in 2D invariance that can be used to advantage
in 3D to 2D projection. We exploit these techniques and model actions in terms of view-invariant canonical body poses and
trajectories in 2D invariance space, leading to a simple and effective way to represent and recognize human actions from a
general viewpoint. We first evaluate the approach theoretically and show why a straightforward application of the 2D invariance
idea will not work. We describe strategies designed to overcome inherent problems in the straightforward approach and outline
the recognition algorithm. We then present results on 2D projections of publicly available human motion capture data as well
on manually segmented real image sequences. In addition to robustness to viewpoint change, the approach is robust enough to
handle different people, minor variabilities in a given action, and the speed of aciton (and hence, frame-rate) while encoding
sufficient distinction among actions.
This work was done when the author was a graduate student in the Department of Computer Science and was partially supported
by the NSF Grant ECS-02-5475. The author is curently with Siemens Corporate Research, Princeton, NJ.
Dr. Chellappa is with the Department of Electrical and Computer Engineering. 相似文献
7.
It is well known that different frames play different roles in feature learning in video based human action recognition task. However, most existing deep learning models put the same weights on different visual and temporal cues in the parameter training stage, which severely affects the feature distinction determination. To address this problem, this paper utilizes the visual attention mechanism and proposes an end-to-end two-stream attention based LSTM network. It can selectively focus on the effective features for the original input images and pay different levels of attentions to the outputs of each deep feature maps. Moreover, considering the correlation between two deep feature streams, a deep feature correlation layer is proposed to adjust the deep learning network parameter based on the correlation judgement. In the end, we evaluate our approach on three different datasets, and the experiments results show that our proposal can achieve the state-of-the-art performance in the common scenarios. 相似文献
8.
Chow BH Chua DT Sham JS Zhang MY Chow LW Bi J Ma NF Xie D Loo WT Fung JM Fu L Guan XY 《Proteomics. Clinical applications》2009,3(6):654-662
Adjuvant chemotherapy alongside radiotherapy is one of the effective therapies in nasopharyngeal carcinoma (NPC) treatment. However, the appearance of drug resistance is a major obstacle for anti-cancer chemotherapy and often causes failure of the chemotherapy. In this study, a drug-resistant gene annexin I (ANX-I) was identified by comparing differentially expressed proteins between a cisplatin (CDDP)-resistant NPC cell line CNE2-CDDP and parental CNE2 cells using 2-DE. When ANX-I was transfected into CNE2 cells, the CDDP resistance of CNE2 cells was dramatically increased. The drug-resistant ability of ANX-I was demonstrated by both in vitro and in vivo assays. The association of ANX-I expression with clinical features was also investigated. Increased expression of ANX-I was significantly associated with disease relapse in NPC (p<0.05). In breast and gastric cancer, increased expression of ANX-I was significantly associated with drug resistance (p<0.001) and poor prognosis (p<0.001), respectively. Taken together, our findings suggest that ANX-I plays an important role in drug resistance. 相似文献
9.
10.
11.
P2PSIP用P2P算法实现SIP的位置管理功能,解决了单点失效问题,有良好的可扩展性,将成为VoIP的关键技术。现有相关研究缺乏对不同网络互通的深入探讨。提出P2PSIP网络与PSTN的一种新颖的互通机制,并且基于该机制设计了一个易于实现的P2PSIP终端与PSTN电话互通的原型系统。互通的关键设计是一个具有双重身份的网关,在P2PSIP网络中作为对等端,在PSTN中负责连接电话机。经测试分析,该互通机制可以应用于实际网络,实现P2PSIP网络和PSTN之间的会话建立。 相似文献
12.
13.
14.
针对现有的人体骨架动作识别方法对肢体信息挖掘不足以及时间特征提取不足的问题,提出了一种基于姿态校正模块与姿态融合模块的模型PTF-SGN,实现了对骨架图关键时空信息的充分利用。首先,对骨架图数据进行预处理,挖掘肢体和关节点的位移信息并提取特征;然后,姿态校正模块通过无监督学习的方式获取姿态调整因子,并对人体姿态进行自适应调整,增强了模型在不同环境下的鲁棒性;其次,提出一种基于时间注意力机制的姿态融合模块,学习骨架图中的短时刻特征与长时刻特征并融合长短时刻特征,加强了对时间特征的表征能力;最后,将骨架图的全局时空特征输入到分类网络中得到动作识别结果。在NTU60 RGB+D、NTU120 RGB+D两个3D骨架数据集和Penn-Action、HARPET两个2D骨架数据集上的实验结果表明,该模型能够有效地识别骨架时序数据的动作。 相似文献
15.
单鹏畅;高利剑;董文龙;毛启容 《计算机工程》2025,51(6):93-101
行为检测任务包含行为分类和边界定位,往往关注行为特征和边界特征。已有方法通常忽略了行为空间特征对于该任务的重要性,并存在行为边界预测模糊的问题,影响行为检测模型的性能和应用效果。针对以上问题,提出一种基于显著目标追踪的行为检测(SOT-AD)方法。首先,为了学习不同尺度的显著空间信息,提出分级注意力网络,旨在捕捉与行为关联的显著目标,减少与行为无关的信息的干扰;其次,为了使相邻时序位置关注到的显著目标具有一致性,提出显著目标追踪损失;最后,引入中性样本辅助构造“目标-次目标-背景”特征池,旨在学习特征时序上下文信息,实现显著目标追踪。在THUMOS14和ActivityNet1.3两个通用数据集上的实验结果表明,与主流方法相比,SOT-AD在平均精度均值(mAP)指标上分别平均提升了0.9和0.6百分点。其中,在THUMOS14数据集上,SOT-AD的mAP@0.5达到72.7%。 相似文献
16.
17.
康复锻炼是脑卒中患者的重要治疗方式,为提高康复动作识别的准确率与实时性,更好地辅助患者在居家环境中进行长期康复训练,结合姿态估计与门控循环单元(GRU)网络提出一种人体康复动作识别算法Pose-AMGRU。采用OpenPose姿态估计方法从视频帧中提取骨架关节点,经过姿态数据预处理后得到表达肢体运动的关键动作特征,并利用注意力机制构建融合三层时序特征的GRU网络实现人体康复动作分类。实验结果表明,该算法在KTH和康复动作数据集中的识别准确率分别为98.14%和100%,且在GTX1060显卡上的运行速度达到14.23frame/s,具有较高的识别准确率与实时性。 相似文献
18.
目的 动作识别在工业生产制造中变得越来越重要。但在复杂的生产车间内,动作识别技术会受到环境遮挡、视角变化以及相似动作识别等干扰。基于此,提出一种结合双视图骨架多流网络的装箱行为识别方法。方法 将堆叠的差分图像(residual frames,RF)作为模型的输入,结合多视图模块解决人体被遮挡的问题。在视角转换模块中,将差分人体骨架旋转到最佳的虚拟观察角度,并将转换后的骨架数据传入3层堆叠的长短时记忆网络(long short-term memory,LSTM)中,将不同视角下的分类分数进行融合,得到识别结果。为了解决细微动作的识别问题,采用结合注意力机制的局部定位图像卷积网络,传入到卷积神经网络中进行识别。融合骨架和局部图像识别的结果,预测工人的行为动作。结果 在实际生产环境下的装箱场景中进行了实验,得到装箱行为识别准确率为92.31%,较大幅度领先于现有的主流行为识别方式。此外,该方法在公共数据集NTU(Nanyang Technological University)RGB+D上进行了评估,结果显示在CS(cross-subject)协议和CV(cross-view)协议中的性能分别达到了85.52%和93.64%,优于其他网络,进一步验证了本文方法的有效性和准确性。结论 本文提出了一种人体行为识别方法,能够充分利用多个视图中的人体行为信息,采用骨架网络和卷积神经网络模型相结合的方式,有效提高了行为识别的准确率。 相似文献
19.
随着流媒体直播应用的日益扩大,传统C/S(Client/Server)模式网络构架的服务器负载过重,导致QoS(Quality of Service)得不到保证,而P2P(Peer-to-Peer)网络具有负载均衡、可扩展性、容错性强等优点。目前P2P技术开始逐步走向成熟。大多数P2P流媒体数据分发采用Pull模式或Push模式,本文提出基于推拉结合机制的P2P流媒体分发算法,通过仿真实验,从系统开销、端到端延迟等方面对两者进行了性能对比。实验结果表明,该推拉结合机制有效的降低了数据块传输时延,并且在很大程度上避免了数据块的重复推送,降低了数据冗余。 相似文献
20.
Clark RD Wolohan PR Hodgkin EE Kelly JH Sussman NL 《Journal of molecular graphics & modelling》2004,22(6):487-497
There is currently a great deal of interest in creating computational tools for predicting the pharmacological properties of drug development candidates, ranging from physicochemical properties such as pKa and solubility to more complex biological properties such as oral bioavailability and toxicity. The limiting factor in many cases is a shortage of good data from which to construct training sets. In other cases, large amounts of data are available, but they use surrogate end-points or are comprised of compounds very different from those usually encountered in drug discovery and development. In such cases large training sets and global models are not necessarily better than local models based on smaller data sets. Such considerations make it as important to examine the available data carefully so as to avoid over-interpretation of the models obtained as it is to minimise errors in prediction per se. The kinds of complications likely to be encountered for in vitro hepatotoxicity modelling are discussed in general terms and illustrated in particular by SIMCA analysis of data obtained from assays of cultured hepatocytes for a large, structurally diverse data set and a smaller, much more focussed one. 相似文献