Motion Style Research Articles

Fast human motion transfer based on a meta network

The paper proposes a new approach to the task of unsupervised human motion style transfer which transforms the style of an unlabeled input human motion into the style of another unlabeled motion sequence and preserves the behavior of the primitive motion at the same time. For supervised stylistic motion transfer task, constructing a database including a large quantity of training sample pairs is expensive and tedious for artists. Therefore, we focus on unsupervised methods. However, recent works need to retrain a network for each new character with different styles, which is a time-consuming and resource-intensive process. To tackle this problem, we utilize a meta network which builds a direct-mapping from the style motion sequence to the transformation network, namely, producing a corresponding motion transformation network in the meta network by a feed-forward propagation. The model transfers multiple styles of motion clips to different motions and generates natural stylized motions in real-time without a re-training for every style. Besides, our model can transfer styles between two movements depending on different skeletons. We also explore important features of the transfer network manifold by interpolation between the hidden states from meta network. Experiments validate the flexibility and effectiveness of our data-driven model.

Unpaired motion style transfer from video to animation

Transferring the motion style from one animation clip to another, while preserving the motion content of the latter, has been a long-standing problem in character animation. Most existing data-driven approaches are supervised and rely on paired data, where motions with the same content are performed in different styles. In addition, these approaches are limited to transfer of styles that were seen during training. In this paper, we present a novel data-driven framework for motion style transfer, which learns from an unpaired collection of motions with style labels, and enables transferring motion styles not observed during training. Furthermore, our framework is able to extract motion styles directly from videos, bypassing 3D reconstruction, and apply them to the 3D input motion. Our style transfer network encodes motions into two latent codes, for content and for style, each of which plays a different role in the decoding (synthesis) process. While the content code is decoded into the output motion by several temporal convolutional layers, the style code modifies deep features via temporally invariant adaptive instance normalization (AdaIN). Moreover, while the content code is encoded from 3D joint rotations, we learn a common embedding for style from either 3D or 2D joint positions, enabling style extraction from videos. Our results are comparable to the state-of-the-art, despite not requiring paired training data, and outperform other methods when transferring previously unseen styles. To our knowledge, we are the first to demonstrate style transfer directly from videos to 3D animations - an ability which enables one to extend the set of style examples far beyond motions captured by MoCap systems.

Pliocene–Pleistocene basin evolution along the Garlock fault zone, Pilot Knob Valley, California

Abstract Exposed Pliocene–Pleistocene terrestrial strata provide an archive of the spatial and temporal development of a basin astride the sinistral Garlock fault in California. In the southern Slate Range and Pilot Knob Valley, an ∼2000-m-thick package of Late Cenozoic strata has been uplifted and tilted to the northeast. We name this succession the formation of Pilot Knob Valley and provide new chronologic, stratigraphic, and provenance data for these rocks. The unit is divided into five members that record different source areas and depositional patterns: (1) the lowest exposed strata are conglomeratic rocks derived from Miocene Eagle Crags volcanic field to the south and east across the Garlock fault; (2) the second member consists mostly of fine-grained rocks with coarser material derived from both southern and northern sources; and (3) the upper three members are primarily coarse-grained conglomerates and sandstones derived from the adjacent Slate Range to the north. Tephrochronologic data from four ash samples bracket deposition of the second member to 3.6–3.3 Ma and the fourth member to between 1.1 and 0.6 Ma. A fifth tephrochronologic sample from rocks south of the Garlock fault near Christmas Canyon brackets deposition of a possible equivalent to the second member of the formation of Pilot Knob Valley at ca. 3.1 Ma. Although the age of the base of the lowest member is not directly dated, regional stratigraphic and tectonic associations suggest that the basin started forming ca. 4–5 Ma. By ca. 3.6 Ma, the northward progradation fanglomerate sourced in the Eagle Crags region waned, and subsequent deposition occurred in shallow lacustrine systems. At ca. 3.3 Ma, southward progradation of conglomerates derived from the Slate Range began. Circa 1.1 Ma, continued southward progradation of fanglomerate with Slate Range sources is characterized by a shift to coarser grain sizes, interpreted to reflect uplift of the Slate Range. Overall, basin architecture and the temporal evolution of different source regions were controlled by activity on three regionally important faults—the Garlock, the Marine Gate, and the Searles Valley faults. The timing and style of motions on these faults appear to be directly linked to patterns of basin development.

Intensive Motion Style Acupuncture Treatment (MSAT) Is Effective for Patients with Acute Whiplash Injury: A Randomized Controlled Trial.

In this single-center, parallel, randomized controlled trial, we aim to examine the effects and safety of motion style acupuncture treatment (MSAT; a combination of acupuncture and Doin therapy) on pain reduction and functional improvement in patients with whiplash-associated disorders (WADs). Ninety-seven patients with cervical pain admitted to the Bucheon Jaseng Hospital of Korean Medicine, South Korea, due to acute whiplash injury were treated with integrative Korean medicine (IKM) with (MSAT group, 48 patients) or without (control group, 49 patients) an additional 3-day MSAT during hospitalization (5–14 days) and followed-up for 90 days. The mean numeric rating scale (NRS) scores of the MSAT and control groups at baseline were 5.67 (95% confidence interval (CI), 5.33, 6.01) and 5.44 (95% CI, 5.06, 5.82), respectively, and on day 5, 3.55 (95% CI, 3.04, 4.06) and 4.59 (95% CI, 4.10–5.07), respectively. The NRS change difference between the groups was −1.07 (95% CI, −1.76, −0.37). The rate of recovery of neck pain (NRS score change ≥ 2 points) was significantly faster in the MSAT than in the control group (log-rank test p = 0.0055). IKM treatment combined with MSAT may be effective in reducing the pain and improving the range of motion in patients with WADs.

Transferring video to dynamic relief structures with water interaction

AbstractIn this article, we present a framework which transfers video to dynamic relief‐structures with water interaction. Our method enables the interaction between video content and waterflows. The key idea is that the dynamic 2.5D structures of the objects in video are constructed and then waterflows are simulated to interact with the 2.5D structures. We performed experiments for still images, and video in which objects had various motion styles. User interaction with our system was also demonstrated. The result of a user study indicated that our method could produce visually acceptable results.

Low-Level Characterization of Expressive Head Motion Through Frequency Domain Analysis

For the purpose of understanding how head motions contribute to the perception of emotion in an utterance, we aim to examine the perception of emotion based on Fourier transform-based static and dynamic features of head motion. Our work is to conduct intra-related objective analysis and perceptual experiments on the link between the perception of emotion and the static/dynamic features. The objective analysis outcome shows that the static and dynamic features are effective in characterizing and recognizing emotions. The perceptual experiments enable us to collect human perception of emotion through head motion. The collected perceptual data shows that humans are unable to reliably perceive emotion from head motion alone but reveals that humans are sensitive to the static feature (in reference to the averaged up-down rotation angle) and the dynamic features (which reflect the fluidity and speed of movement). It also indicates that humans perceive emotion carried in head motion and the naturalness of head motion in two different channels. Our work contributes to the understanding and the characterization of head motion in expressive speech through low-level descriptions of motion features, instead of commonly used high-level motion style (e.g., head nods, shakes, tilts, and raises).

A Case Report of Complex Korean Medicine Treatments for Tetraplegia Caused by Spinal Cord Injury

Background This study aimed to determine the effects of complex Korean medicine treatments on a patient suffering from tetraplegia caused by spinal cord injury. Case Summary A 64-year-old female patient diagnosed with tetraplegia was treated using acupuncture, electroacupuncture, pharmacopuncture, knee joint motion style treatment (MST), and herbal medicine. Clinical symptoms were measured using the numeric rating scale (NRS) and manual muscle test (MMT). After 67 days of treatment, the NRS score for lower back pain and knee pain decreased from 7 to 3. For the shoulder joint the MMT grade improved from 3+/4+(Rt./Lt.) to 4/5-; in the hip joint, it improved from 3-/3+ to 4/4+. No side effects were observed from the treatments used in this case report. Conclusion The complex Korean Medicine treatments appeared to be effective in recovering muscle strength and reducing pain in patients with tetraplegia caused by spinal cord injury. Keywords: tetraplegia, spinal cord injury, motion style treatment, pharmacopuncture, Chungpajunsin-bang

Construction of Virtual Video Scene and Its Visualization During Sports Training

This article studies the actual captured human motion data for human motion synthesis and style transfer, constructs a scene of motion virtual video, and attempts to directly generate human motion style video to establish a sports style transfer model that combines and self-encoding. The original human motion capture data mapped to the motion feature space for style transfer synthesis. The coding network used to map the high-dimensional motion capture data to the low-dimensional feature space, and the motion style transfer constraints established in the feature space, and the human body motion results after the style transfer obtained by decoding. This paper proposes a pixel-level human motion style transfer model based on conditional adversarial networks and uses convolution and convolution to establish two branch coding networks to extract the features of the input style video and content pictures. The decoding network decodes the combined two features and generates a human motion video data frame by frame. The Gram matrix establishes constraints on the encoding and decoding features, controls the movement style of the human body, and finally realizes the visualization of the movement process. The incremental learning method based on the cascade network can improve the high accuracy and achieve the posture measurement frequency of 200Hz. The research results provide a key foundation for improving the immersion sensation of sport visual and tactile interaction simulation.

Survey on Style in 3D Human Body Motion: Taxonomy, Data, Recognition and Its Applications

The meaning of the word style depends on its context. While actions have already been quite studied for a while, style in human body motion is a growing topic of interest. In the context of animation, style is crucial as it brings realism and expressiveness to the motion of a character. Even though it is undoubtedly a key element in motions, its definition and the use of the word style in itself, among research works, lack consensus. Achieving realistic motions is tedious. It requires either a large motion capture dataset or the considerable work of artist animators. The lack of consistent style data is thus a challenge. Stylistic motion generation is quite studied in order to overcome this issue. This paper focuses on the study of style in human body motion from 3D human body skeletal data. It establishes a taxonomy of definitions of style, describes the data that have been used up until now, introduces key notions about motion capture data as well as machine learning, and presents approaches about style recognition, person identification through their style and motion style generation.

Neural style-preserving visual dubbing

Dubbing is a technique for translating video content from one language to another. However, state-of-the-art visual dubbing techniques directly copy facial expressions from source to target actors without considering identity-specific idiosyncrasies such as a unique type of smile. We present a style-preserving visual dubbing approach from single video inputs, which maintains the signature style of target actors when modifying facial expressions, including mouth motions, to match foreign languages. At the heart of our approach is the concept of motion style, in particular for facial expressions, i.e., the person-specific expression change that is yet another essential factor beyond visual accuracy in face editing applications. Our method is based on a recurrent generative adversarial network that captures the spatiotemporal co-activation of facial expressions, and enables generating and modifying the facial expressions of the target actor while preserving their style. We train our model with unsynchronized source and target videos in an unsupervised manner using cycle-consistency and mouth expression losses, and synthesize photorealistic video frames using a layered neural face renderer. Our approach generates temporally coherent results, and handles dynamic backgrounds. Our results show that our dubbing approach maintains the idiosyncratic style of the target actor better than previous approaches, even for widely differing source and target actors.

Stylized human motion warping method based on identity-independent coordinates

Human motion style is a vital concept to virtual human that has a great impact on the expressiveness of the final animation. This paper presents a novel technique that transfers style between heterogeneous motions in real time. Unlike previous approaches, our stylized motion warping is capable of reusing style between heterogeneous motions. The key idea of our work is to represent human motions with identity-independent coordinates (IICs) and learn relative space–time transformations between stylistically different IICs, instead of separating style from content. Once the relative space–time transformations are estimated from a small set of stylized example motions whose contents are identical, our technique is capable of generating style-controllable human motions by applying these transformations to heterogeneous motions with simple linear operations. Experimental results demonstrate that our technique is efficient and powerful in stylized human motion generation. Besides, our technique can also be used in numerous interactive applications, such as real-time human motion style control, stylizing motion graphs and style-based human motion editing.

Exaggerated Cartoon Style Motion in Hotel Transylvania

The fundamental principles of animation were developed at the Disney studio in the 1930s and when applied to animation result in a recognisable “Disney” style of motion. By adhering to these fundamental principles of animation, Animators have been able to successfully extend the Disney style of motion to the 3D CG medium. This paper goes beyond Disney style motion and examines the adaptation of techniques typically seen in more extreme cartoon-style motion and “pushed posing”, exemplified by early Warner Brothers animated short films. This is achieved via an analysis of Hotel Transylvania by Sony Pictures Animation, which features a stylised form of animation adapted from the extended principles of extreme cartoon-style motion to the 3D medium.

KREATIVITAS DWIYASMONO DALAM KARYA TARI HARJUNA KERATARUPA

This research focuses on Dwiyasmono’s creativity in creating Harjuna–Keratarupo Dance. Harjuna–Keratarupo Dance is a phetilan dance two adult males. Issues raised in the research is about the form of Harjuna–Keratarupo Dance. And about Dwiyasmono’s creativity in the Creations of Harjuna–Keratarupo Dance. This research is a Qualitative research using descriptive methods of analysis and the technical collection of observations, interviews, bibliography studies and Documents. To analyse the shape of the dish using form concepts expressed by Suzane K. Langer and its elements the constituent was outlined using Soedarsono’s theory. To Explain Dwiyasmono’s creativity analyzed using theory Rhodes that creativity there are four elements, namely person, press, process, product. The results of the research known as Harjuna-Keratarupo Dance in the form of a divided into three parts, namely the Maju Beksan, Beksan and Mundur Beksan. This dance has the merging of two styles of dance motion. Dwiyasmono’s creativity in the cupping section of the Surakarta movement is combined with the movement of Yogyakarta style. The dance music in the dance is colored with a leather puppet pack. The development of the garage is influenced by external factors and internal factors, especially for its withdrawal.Keywords: Harjuna Keratarupo, form, Dwiyasmono, creativity.

Human motion recognition based on Kalman random Forest algorithm and 3D multimedia

A human motion recognition method based on Kalman random forest algorithm model is proposed to improve the accuracy and efficiency of tracking algorithm aiming at the multi-degree-of-freedom problem of traditional human motion recognition and tracking. Firstly, a simple explicit manifold space model of human motion parameters is established. Reconstruction of three-dimensional image features for human motion requires accurate identification of human behavioral features. The human motion process is highly random, and the sudden change in motion amplitude causes a sudden change in the number of shape-based parameters in the three-dimensional feature. The compact low-dimensional model simplifies the difficulty of human motion parameters contour tracking; Secondly, Kalman random forest algorithm is used to estimate the global transformation of action and the different distance measures of body action in each sub-state by using action characteristics through boundary approximation and action sequence updating, so as to reduce the computational complexity of the algorithm; Finally, the single shape change and the dynamic action in motion are modeled, and the dynamic tracking of human motion is realized through the linear combination of human motion shape and style factor decomposition.

Efficient Neural Networks for Real-time Motion Style Transfer

Style is an intrinsic, inescapable part of human motion. It complements the content of motion to convey meaning, mood, and personality. Existing state-of-the-art motion style methods require large quantities of example data and intensive computational resources at runtime. To ensure output quality, such style transfer applications are often run on desktop machine with GPUs and significant memory. In this paper, we present a fast and expressive neural network-based motion style transfer method that generates stylized motion with quality comparable to the state of the art method, but uses much less computational power and a much smaller memory footprint. Our method also allows the output to be adjusted in a latent style space, something not offered in previous approaches. Our style transfer model is implemented using three multi-layered networks: a pose network, a timing network and a foot-contact network. A one-hot style vector serves as an input control knob and determines the stylistic output of these networks. During training, the networks are trained with a large motion capture database containing heterogeneous actions and various styles. Joint information vectors together with one-hot style vectors are extracted from motion data and fed to the networks. Once the network has been trained, the database is no longer needed on the device, thus removing the large memory requirement of previous motion style methods. At runtime, our model takes novel input and allows real-valued numbers to be specified in the style vector, which can be used for interpolation, extrapolation or mixing of styles. With much lower memory and computational requirements, our networks are efficient and fast enough for real-time use on mobile devices. Requiring no information about future states, the style transfer can be performed in an online fashion. We validate our result both quantitatively and perceptually, confirming its effectiveness and improvement over previous approaches.

제한된 모션 센서와 애니메이션 데이터를 이용한 캐릭터 동작 제어

A 3D virtual character playing a role in a digital story-telling has a unique style in its appearance and motion. Because the style reflects the unique personality of the character, it is very important to preserve the style and keep its consistency. However, when the character’s motion is directly controlled by a user’s motion who is wearing motion sensors, the unique style can be discarded. We present a novel character motion control method that uses only a small amount of animation data created only for the character to preserve the style of the character motion. Instead of machine learning approaches requiring a large amount of training data, we suggest a search-based method, which directly searches the most similar character pose from the animation data to the current user’s pose. To show the usability of our method, we conducted our experiments with a character model and its animation data created by an expert designer for a virtual reality game. To prove that our method preserves well the original motion style of the character, we compared our result with the result obtained by using general human motion capture data. In addition, to show the scalability of our method, we presented experimental results with different numbers of motion sensors.

CHOREOGRAPHIC WORKS AS OBJECTS OF FORENSIC EXPERTISE: KEY FEATURES AND APPROACHES TO SOLVING EXPERT TASKS

A research aim was to investigate signs peculiar to choreographic works, as objects of copyright. Will consider, what elements of choreographic results can be attributed to maintenance, and also internal and external forms, and also to offer approach to the decision of set certain tasks. Identification of the main features of choreographic works has a special relevance for the methodological provision of forensic expert activity. It was established that the main tasks of the forensic expertise of choreographic works are: the definition of the object of copyright law in the object of research; definition in the object of the research or its part of signs of originality and creative nature; establishing the fact of reproduction of a choreographic work; establishing a method for its use or processing. The results of the conducted research indicate that the elements of the internal form of choreographic works may include the artistic image, composition, structure of the work. The elements of the external form – the choreographic work of the work (choreographic vocabulary of the work, style of speech), that is, the author’s own means and techniques of creating artistic images, with which plastic dance images are created. During realization of expertise it is expedient to investigate the next elements of choreographic language works: composition of motions, gestures, poses, mimicry; harmoniousness of motions and poses, plastic expressiveness and mimicry, rate and rhythm of motion, spatial picture; light and colour decisions; stage sceneries, suits, make-up and hair-dos of actors; individual style of motion (handwriting of motions) etc. At establishment of fact of the complete or partial use of choreographic work an expert on the stage of comparative research conducts the analysis of language elements of the investigated works, methods of their selection, use, mutual combination and correlation. The elements of maintenance and form of these works are thus analysed, in particular, features of expression of their internal and external form. In the process of research the special attention applies on facts that can testify to more early use by other authors of analogical choreographic elements. It also establishes the originality of the used choreographic vocabulary of the work, means and methods of creating artistic images. Key words: choreographic works, forms of work, expert practice, objects of copyright.

Local interaction rules and collective motion in black neon tetra (Hyphessobrycon herbertaxelrodi) and zebrafish (Danio rerio).

We explored the local motion rules used by interacting individuals in small groups of black neon tetra (Hyphessobrycon herbertaxelrodi) and zebrafish (Danio rerio) to ascertain if and how these rules underlie the fishes' global collective coordinated motion. As these 2 species show very different styles of collective motion in terms of cohesion and polarization, we expected to find differences in their individual behavioral rules. We recorded groups of 2, 3, 4, and 8 fish of each species; tracked their individual trajectories; and studied how their individual turning angles and accelerations varied as a function of heading differences, distances, and relative angles to their neighbors. We found that black neon tetra and zebrafish differed in terms of their preferential positions with respect to their neighbors, the magnitude of turning angles and accelerations, and the way these angles and accelerations are modulated by both the distance from neighbors (thus suggesting a "repulsion" zone in black neon tetra but not in zebrafish) and the heading difference and relative angle to neighbors. Our results enable us to infer that, in black neon tetra, avoiding excessive proximity and collision takes priority over cohesion, and cohesion takes priority over polarization. This provides evidence that rules are similar in species of very different genera and that differences are a matter of degree. Our results also provide substantial empirical evidence to support the theoretical assumptions made in agent-based models that simulate coordinated collective motion in many different animal species. (PsycINFO Database Record (c) 2019 APA, all rights reserved).

Laban-Inspired Task-Constrained Variable Motion Generation on Expressive Aerial Robots

This paper presents a method for creating expressive aerial robots through an algorithmic procedure for creating variable motion under given task constraints. This work is informed by the close study of the Laban/Bartenieff movement system, and movement observation from this discipline will provide important analysis of the method, offering descriptive words and fitting contexts—a choreographic frame—for the motion styles produced. User studies that use utilize this qualitative analysis then validate that the method can be used to generate appropriate motion in in-home contexts. The accuracy of an individual descriptive word for the developed motion is up to 77% and context accuracy is up to 83%. A capacity for state discernment from motion profile is essential in the context of projects working toward developing in-home robots.

Communicating Dominance in a Nonanthropomorphic Robot Using Locomotion

Dominance is a key aspect of interpersonal relationships. To what extent do nonverbal indicators related to dominance status translate to a nonanthropomorphic robot? An experiment ( N = 25) addressed whether a mobile robot's motion style can influence people's perceptions of its status. Using concepts from improv theater literature, we developed two motion styles across three scenarios (robot makes lateral motions, approaches, and departs) to communicate a robot's dominance status through nonverbal expression. In agreement with the literature, participants described a motion style that was fast, in the foreground, and more animated as higher status than a motion style that was slow, in the periphery, and less animated. Participants used fewer negative emotion words to describe the robot with the purportedly high-status movements versus the purportedly low-status movements, but used more negative emotion words to describe the robot when it made departing motions that occurred in the same style. This result provides evidence that guidelines from improvisational theater for using nonverbal expression to perform interpersonal status can be applied to influence perception of a nonanthropomorphic robot's status, thus suggesting that useful models for more complicated behaviors might similarly be derived from performance literature and theory.