scholarly journals Learning Diverse Stochastic Human-Action Generators by Learning Smooth Latent Transitions

2020 ◽  
Vol 34 (07) ◽  
pp. 12281-12288
Author(s):  
Zhenyi Wang ◽  
Ping Yu ◽  
Yang Zhao ◽  
Ruiyi Zhang ◽  
Yufan Zhou ◽  
...  
Keyword(s):  
Human Action ◽  
Human Motion ◽  
Motion Generation ◽  
Action Sequence ◽  
Sequence Generation ◽  
Latent Space ◽  
Action Sequences ◽  
Mixed Classes ◽  
Standard Action ◽  
Net Framework

Human-motion generation is a long-standing challenging task due to the requirement of accurately modeling complex and diverse dynamic patterns. Most existing methods adopt sequence models such as RNN to directly model transitions in the original action space. Due to high dimensionality and potential noise, such modeling of action transitions is particularly challenging. In this paper, we focus on skeleton-based action generation and propose to model smooth and diverse transitions on a latent space of action sequences with much lower dimensionality. Conditioned on a latent sequence, actions are generated by a frame-wise decoder shared by all latent action-poses. Specifically, an implicit RNN is defined to model smooth latent sequences, whose randomness (diversity) is controlled by noise from the input. Different from standard action-prediction methods, our model can generate action sequences from pure noise without any conditional action poses. Remarkably, it can also generate unseen actions from mixed classes during training. Our model is learned with a bi-directional generative-adversarial-net framework, which can not only generate diverse action sequences of a particular class or mix classes, but also learns to classify action sequences within the same model. Experimental results show the superiority of our method in both diverse action-sequence generation and classification, relative to existing methods.

scholarly journals Temporal-order judgement task suggests chronological action representations in motor experts and non-experts

2020 ◽  
Vol 73 (11) ◽  
pp. 1879-1890
Author(s):  
Róisín Elaine Harrison ◽  
Martin Giesel ◽  
Constanze Hesse
Keyword(s):  
Temporal Order ◽  
Human Action ◽  
Natural Order ◽  
Action Sequence ◽  
Motor Expertise ◽  
Motor Priming ◽  
Human Movements ◽  
Action Sequences ◽  
Temporal Order Judgement ◽  
Tentative Evidence

Motor priming studies have suggested that human movements are mentally represented in the order in which they usually occur (i.e., chronologically). In this study, we investigated whether we could find evidence for these chronological representations using a paradigm which has frequently been employed to reveal biases in the perceived temporal order of events—the temporal-order judgement task. We used scrambled and unscrambled images of early and late movement phases from an everyday action sequence (“stepping”) and an expert action sequence (“sprinting”) to examine whether participants’ mental representations of actions would bias their temporal-order judgements. In addition, we explored whether motor expertise mediated the size of temporal-order judgement biases by comparing the performances of sprinting experts with those of non-experts. For both action types, we found significant temporal-order judgement biases for all participants, indicating that there was a tendency to perceive images of human action sequences in their natural order, independent of motor expertise. Although there was no clear evidence that sprinting experts showed larger biases for sprinting action sequences than non-experts, considering sports expertise in a broader sense provided some tentative evidence for the idea that temporal-order judgement biases may be mediated by more general motor and/or perceptual familiarity with the running action rather than specific motor expertise.

Guided latent space regression for human motion generation

2013 ◽  
Vol 61 (4) ◽  
pp. 340-350 ◽  
Author(s):  
Carlo Alberto Avizzano
Keyword(s):  
Human Motion ◽  
Motion Generation ◽  
Latent Space

scholarly journals Independence Action Recognition using Self Similarities

2019 ◽  
Vol 8 (12) ◽  
pp. 5342-5346
Keyword(s):  
Action Recognition ◽  
Human Action Recognition ◽  
Human Action ◽  
Geometric Properties ◽  
Action Sequence ◽  
Action Sequences ◽  
Autonomous Activity ◽  
Similar Class ◽  
Dynamic Time ◽  
Over Time

Exploring Self-Similarities Of Action Sequences Over Time And Observing The Striking Stability Of Human Action Recognition. Developing An Action Descriptor That Captures The Structure Of Temporal Similarities And Dissimilarities Within An Action Sequence. Self-Likeness Descriptors Are Demonstrated To Be Steady Under Execution Varieties Inside A Group Of Activities When Person Haste Changes Are Overlooked. Changes Between Two Unique Occurrences Of A Similar Class Can Be Unequivocally Recouped With Dynamic Time Traveling. Adequate Activity Separations Are As Yet Held Along These Lines To Construct A View-Autonomous Activity Acknowledgment Framework. Strangely Self-Likenesses Are Registered From Various Picture Highlights Have Comparable Properties And Can Be Utilized In A Corresponding Manner. It Depends Powerless Geometric Properties And Joins Them With AI For Proficient Cross-See Activity Acknowledgment. It Has Comparative Or Better Execution Looked At Than Related Techniques And It Performs Well Even In Outrageous Conditions, For Example, Well Perceiving Activities From Top Perspectives.

scholarly journals The role of action effects in motor sequence planning and execution: exploring the influence of temporal and spatial effect anticipation

2021 ◽  
Author(s):  
Rachel M. Brown ◽  
Erik Friedgen ◽  
Iring Koch
Keyword(s):  
Action Planning ◽  
Spatial Effects ◽  
Motor Sequence ◽  
Action Sequence ◽  
Sequential Action ◽  
Spatial Features ◽  
Temporal Features ◽  
Action Sequences ◽  
Temporal And Spatial ◽  
Anticipation Effect

AbstractActions we perform every day generate perceivable outcomes with both spatial and temporal features. According to the ideomotor principle, we plan our actions by anticipating the outcomes, but this principle does not directly address how sequential movements are influenced by different outcomes. We examined how sequential action planning is influenced by the anticipation of temporal and spatial features of action outcomes. We further explored the influence of action sequence switching. Participants performed cued sequences of button presses that generated visual effects which were either spatially compatible or incompatible with the sequences, and the spatial effects appeared after a short or long delay. The sequence cues switched or repeated across trials, and the predictability of action sequence switches was varied across groups. The results showed a delay-anticipation effect for sequential action, whereby a shorter anticipated delay between action sequences and their outcomes speeded initiation and execution of the cued action sequences. Delay anticipation was increased by predictable action switching, but it was not strongly modified by the spatial compatibility of the action outcomes. The results extend previous demonstrations of delay anticipation to the context of sequential action. The temporal delay between actions and their outcomes appears to be retrieved for sequential planning and influences both the initiation and the execution of actions.

scholarly journals Study on Machine Learning and Deep Learning Methods for Human Action Recognition

2020 ◽  
Author(s):  
Gopika Rajendran ◽  
Ojus Thomas Lee ◽  
Arya Gopi ◽  
Jais jose ◽  
Neha Gautham
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Action Recognition ◽  
Human Body ◽  
Human Performance ◽  
Human Action Recognition ◽  
Human Action ◽  
Human Robot Interaction ◽  
Learning Approach ◽  
Action Sequence

With the evolution of computing technology in many application like human robot interaction, human computer interaction and health-care system, 3D human body models and their dynamic motions has gained popularity. Human performance accompanies human body shapes and their relative motions. Research on human activity recognition is structured around how the complex movement of a human body is identified and analyzed. Vision based action recognition from video is such kind of tasks where actions are inferred by observing the complete set of action sequence performed by human. Many techniques have been revised over the recent decades in order to develop a robust as well as effective framework for action recognition. In this survey, we summarize recent advances in human action recognition, namely the machine learning approach, deep learning approach and evaluation of these approaches.

scholarly journals Using a Multilearner to Fuse Multimodal Features for Human Action Recognition

2020 ◽  
Vol 2020 ◽  
pp. 1-18
Author(s):  
Chao Tang ◽  
Huosheng Hu ◽  
Wenjian Wang ◽  
Wei Li ◽  
Hua Peng ◽  
...  
Keyword(s):  
Action Recognition ◽  
Feature Fusion ◽  
Human Action Recognition ◽  
Human Action ◽  
Human Motion ◽  
Image Features ◽  
Depth Image ◽  
Good Ability ◽  
Multimodal Features ◽  
Feature Based

The representation and selection of action features directly affect the recognition effect of human action recognition methods. Single feature is often affected by human appearance, environment, camera settings, and other factors. Aiming at the problem that the existing multimodal feature fusion methods cannot effectively measure the contribution of different features, this paper proposed a human action recognition method based on RGB-D image features, which makes full use of the multimodal information provided by RGB-D sensors to extract effective human action features. In this paper, three kinds of human action features with different modal information are proposed: RGB-HOG feature based on RGB image information, which has good geometric scale invariance; D-STIP feature based on depth image, which maintains the dynamic characteristics of human motion and has local invariance; and S-JRPF feature-based skeleton information, which has good ability to describe motion space structure. At the same time, multiple K-nearest neighbor classifiers with better generalization ability are used to integrate decision-making classification. The experimental results show that the algorithm achieves ideal recognition results on the public G3D and CAD60 datasets.

scholarly journals Projection Analysis Optimization for Human Transition Motion Estimation

2019 ◽  
Vol 2019 ◽  
pp. 1-9
Author(s):  
Wanyi Li ◽  
Feifei Zhang ◽  
Qiang Chen ◽  
Qian Zhang
Keyword(s):  
Video Game ◽  
Human Motion ◽  
High Dimensional ◽  
Dynamical Models ◽  
Excellent Performance ◽  
3 Dimensional ◽  
Latent Space ◽  
Projection Analysis ◽  
Low Dimensional ◽  
Working Efficiency

It is a difficult task to estimate the human transition motion without the specialized software. The 3-dimensional (3D) human motion animation is widely used in video game, movie, and so on. When making the animation, human transition motion is necessary. If there is a method that can generate the transition motion, the making time will cost less and the working efficiency will be improved. Thus a new method called latent space optimization based on projection analysis (LSOPA) is proposed to estimate the human transition motion. LSOPA is carried out under the assistance of Gaussian process dynamical models (GPDM); it builds the object function to optimize the data in the low dimensional (LD) space, and the optimized data in LD space will be obtained to generate the human transition motion. The LSOPA can make the GPDM learn the high dimensional (HD) data to estimate the needed transition motion. The excellent performance of LSOPA will be tested by the experiments.

Optimal action sequence generation for assistive agents in fixed horizon tasks

2020 ◽  
Vol 34 (2) ◽  
Author(s):  
Kim Baraka ◽  
Francisco S. Melo ◽  
Marta Couto ◽  
Manuela Veloso
Keyword(s):  
Action Sequence ◽  
Sequence Generation ◽  
Optimal Action

A Novel Application of Levenshtein Distance for Assessment of High-Level Motor Planning Underlying Performance During Learning of Complex Motor Sequences

2020 ◽  
Vol 8 (1) ◽  
pp. 67-86 ◽  
Author(s):  
Theresa C. Hauge ◽  
Garrett E. Katz ◽  
Gregory P. Davis ◽  
Kyle J. Jaquess ◽  
Matthew J. Reinhard ◽  
...  
Keyword(s):  
Mental Workload ◽  
Motor Planning ◽  
Reference Sequence ◽  
Levenshtein Distance ◽  
Complex Action ◽  
Action Sequence ◽  
Motor Efficiency ◽  
Practice Performance ◽  
Action Sequences ◽  
High Level

Few studies have examined high-level motor plans underlying cognitive-motor performance during practice of complex action sequences. These investigations have assessed performance through fairly simple metrics without examining how practice affects the structures of action sequences. By adapting the Levenshtein distance (LD) method to the motor domain, we propose a computational approach to accurately capture performance dynamics during practice of action sequences. Practice performance dynamics were assessed by computing the LD based on the number of insertions, deletions, and substitutions of actions needed to transform any sequence into a reference sequence (having a minimal number of actions to complete the task). Also, combining LD-based performance with mental workload metrics allowed assessment of cognitive-motor efficiency dynamics. This approach was tested on the Tower of Hanoi task. The findings revealed that throughout practice this method could capture: i) action sequence performance improvements as indexed by a reduced LD (decrease of insertions and substitutions), ii) structural modifications of the high-level plans, iii) an attenuation of mental workload, and iv) enhanced cognitive-motor efficiency. This effort complements prior work examining the practice of complex action sequences in healthy adults and has potential for probing cognitive-motor impairment in clinical populations as well as the development/assessment of cognitive robotic controllers.

Sign in / Sign up

Export Citation Format

Share Document