Hasil untuk "Dancing"

R. Kanter

David M. Garner, P. Garfinkel

Philip A. Dennis, E. Hall

Haichao Wang, Alexander Okupnik, Yuxing Han et al.

Long-range human movement generation remains a central challenge in computer vision and graphics. Generating coherent transitions across semantically distinct motion domains remains largely unexplored. This capability is particularly important for applications such as dance choreography, where movements must fluidly transition across diverse stylistic and semantic motifs. We propose a simple and effective inference-time optimization framework inspired by diffusion-based stochastic optimal control. Specifically, a control-energy objective that explicitly regularizes the transition trajectories of a pretrained diffusion model. We show that optimizing this objective at inference time yields transitions with fidelity and temporal coherence. This is the first work to provide a general framework for controlled long-range human motion generation with explicit transition modeling.

Pierpaolo Lagani

The dancing venue La Bussola, built in Marina di Pietrasanta between 1946 and 1948 by Baroni, Tempestini, and Porcinai, is an emblematic case of post-war leisure architecture. Originally conceived as a seaside facility, it quickly became a symbol of Italy’s entertainment society under Sergio Bernardini, gaining immediate cultural and social recognition. Architecturally, however, the building underwent radical transformations that compromised its original layout, allowing function to prevail over form and preventing an effective process of heritagization. Through archival research, this paper reconstructs the design phases and highlights the modernity of the original project, while underlining the paradox of its progressive alteration. The case demonstrates how historical knowledge remains a necessary condition for safeguarding the twentieth-century heritage.

Chaozhuo Li, Pengbo Wang, Chenxu Wang et al.

Edgar Allan Poe noted, "Truth often lurks in the shadow of error," highlighting the deep complexity intrinsic to the interplay between truth and falsehood, notably under conditions of cognitive and informational asymmetry. This dynamic is strikingly evident in large language models (LLMs). Despite their impressive linguistic generation capabilities, LLMs sometimes produce information that appears factually accurate but is, in reality, fabricated, an issue often referred to as 'hallucinations'. The prevalence of these hallucinations can mislead users, affecting their judgments and decisions. In sectors such as finance, law, and healthcare, such misinformation risks causing substantial economic losses, legal disputes, and health risks, with wide-ranging consequences.In our research, we have methodically categorized, analyzed the causes, detection methods, and solutions related to LLM hallucinations. Our efforts have particularly focused on understanding the roots of hallucinations and evaluating the efficacy of current strategies in revealing the underlying logic, thereby paving the way for the development of innovative and potent approaches. By examining why certain measures are effective against hallucinations, our study aims to foster a comprehensive approach to tackling this issue within the domain of LLMs.

Clarice Hilton, Kat Hawkins, Phill Tew et al.

Motion capture technologies are increasingly used in creative and performance contexts but often exclude disabled practitioners due to normative assumptions in body modeling, calibration, and avatar representation. EqualMotion introduces a body-agnostic, wearable motion capture system designed through a disability-centred co-design approach. By enabling personalised calibration, integrating mobility aids, and adopting an inclusive visual language, EqualMotion supports diverse body types and movement styles. The system is developed collaboratively with disabled researchers and creatives, aiming to foster equitable participation in digital performance and prototyping. This paper outlines the system's design principles and highlights ongoing case studies in dance and music to evaluate accessibility in real-world creative workflows.

Yuelin Guo, Haoyu He, Zhiyuan Chen et al.

Weakly supervised object detection (WSOD) has attracted significant attention in recent years, as it does not require box-level annotations. State-of-the-art methods generally adopt a multi-module network, which employs WSDDN as the multiple instance detection network module and uses multiple instance refinement modules to refine performance. However, these approaches suffer from three key limitations. First, existing methods tend to generate pseudo GT boxes that either focus only on discriminative parts, failing to capture the whole object, or cover the entire object but fail to distinguish between adjacent intra-class instances. Second, the foundational WSDDN architecture lacks a crucial background class representation for each proposal and exhibits a large semantic gap between its branches. Third, prior methods discard ignored proposals during optimization, leading to slow convergence. To address these challenges, we propose the Dual-thresholded heAtmap-guided proposal clustering and Negative Certainty supervision with Enhanced base network (DANCE) method for WSOD. Specifically, we first devise a heatmap-guided proposal selector (HGPS) algorithm, which utilizes dual thresholds on heatmaps to pre-select proposals, enabling pseudo GT boxes to both capture the full object extent and distinguish between adjacent intra-class instances. We then construct a weakly supervised basic detection network (WSBDN), which augments each proposal with a background class representation and uses heatmaps for pre-supervision to bridge the semantic gap between matrices. At last, we introduce a negative certainty supervision (NCS) loss on ignored proposals to accelerate convergence. Extensive experiments on the challenging PASCAL VOC and MS COCO datasets demonstrate the effectiveness and superiority of our method. Our code is publicly available at https://github.com/gyl2565309278/DANCE.

Alessio Kandiah, Alexander B. Movchan, Vladimir Frid

A new modelling approach shows how the Earth's hidden vibrations may drive global weather dynamics and atmospheric pressure variations, hinting that the planet's own beat could be imprinted on our climate. The atmospheric rotational patterns of the mean sea level pressure, in connection to the development of powerful storms, are shown to be caused by Earth's rotational elastic dynamics and earthquake-induced oscillations. These seismic excitations are discussed in relation to storm formation and the global atmospheric patterns of high-pressure regions.

Hugo Levy-Falk, Suman Sarkar, Thanh Trung Huynh et al.

This study explores the organization of graphene quantum dots on the surface of monocrystalline halide perovskite. We show that graphene quantum dots tends to aggregate on the surface of perovskite unlike in solution or on other substrates, even at very low concentration of the initial solution that should yield single-molecule samples. Spectral analysis on small clusters shows a back-and-forth dynamical transition between an uncoupled, monomer-like state, and an excimer state. Following this "dance" between states, a drastic one-way increase in fluorescence intensity combined with a shortening of the excited state lifetime has been observed on some clusters. This behavior is related to the emission of a collective state that may be a consequence of the dynamical organization of graphene quantum dots under illumination on the surface of the perovksite.

Yue Feng, Jinwei Hu, Qijia Lu et al.

We propose the Multi-modal Untrimmed Video Retrieval task, along with a new benchmark (MUVR) to advance video retrieval for long-video platforms. MUVR aims to retrieve untrimmed videos containing relevant segments using multi-modal queries. It has the following features: 1) Practical retrieval paradigm: MUVR supports video-centric multi-modal queries, expressing fine-grained retrieval needs through long text descriptions, video tag prompts, and mask prompts. It adopts a one-to-many retrieval paradigm and focuses on untrimmed videos, tailored for long-video platform applications. 2) Multi-level visual correspondence: To cover common video categories (e.g., news, travel, dance) and precisely define retrieval matching criteria, we construct multi-level visual correspondence based on core video content (e.g., news events, travel locations, dance moves) which users are interested in and want to retrieve. It covers six levels: copy, event, scene, instance, action, and others. 3) Comprehensive evaluation criteria: We develop 3 versions of MUVR (i.e., Base, Filter, QA). MUVR-Base/Filter evaluates retrieval models, while MUVR-QA assesses MLLMs in a question-answering format. We also propose a Reranking Score to evaluate the reranking ability of MLLMs. MUVR consists of 53K untrimmed videos from the video platform Bilibili, with 1,050 multi-modal queries and 84K matches. Extensive evaluations of 3 state-of-the-art video retrieval models, 6 image-based VLMs, and 10 MLLMs are conducted. MUVR reveals the limitations of retrieval methods in processing untrimmed videos and multi-modal queries, as well as MLLMs in multi-video understanding and reranking. Our code and benchmark is available at https://github.com/debby-0527/MUVR.

Sauptik Dhar, Naveen Ramakrishnan, Michelle Munson

Large Vision Language models have seen huge application in several sports use-cases recently. Most of these works have been targeted towards a limited subset of popular sports like soccer, cricket, basketball etc; focusing on generative tasks like visual question answering, highlight generation. This work analyzes the applicability of the modern video foundation models (both encoder and decoder) for a very niche but hugely popular dance sports - breakdance. Our results show that Video Encoder models continue to outperform state-of-the-art Video Language Models for prediction tasks. We provide insights on how to choose the encoder model and provide a thorough analysis into the workings of a finetuned decoder model for breakdance video classification.

Wojciech Kupczak, Elżbieta Szymańska, Jakub Kozłowski et al.

Background: Undoubtedly, in all areas of physical culture, dance has an extremely positiveimpact on the formation of virtually all motor skills - conditioning, coordination, also affectsagility, speed and flexibility. The aim of this study was to evaluate the level of training loadsin a six-week mesocycle during the starting period of athletes of Latin American style sportballroom dancing and to study the effect of training loads on body fat and energyexpenditure.Methods: The study group consisted of 4 sport-active dance couples with about 12 years ofexperience. Sport-testers (POLAR H7) were used to evaluate training loads. training logsand a body composition analyzer. Results were compared between men and women.Results: The men's group showed higher values of the following training indicators: averagetraining time, energy expenditure, average heart rate. Statistically significant differencesbetween male and female dancers showed such factors as average energy expenditure,average percentage of work in the fifth zone of exercise intensity - in favor of men whileaverage percentage of fat burned - in favor of women.Conclusion: Based on the analysis of energy expenditure and with regard to theclassification of work severity, the dancers performed heavy and very heavy work during thecompetition period, and spent the vast majority of their preparation time in aerobic zones.Accordingly, coaches of Latin American dancers in their own work did not take into accountthe specific physiological workloads performed by them during dance competitions.

Diana-Lidia Tache-Codreanu, Andrei Tache-Codreanu

Art therapy is employed in numerous ways in rehabilitation. This study focuses on an art and movement therapy project carried out during the COVID-19 pandemic. Acting and dancing methods were adapted to produce a short musical film series for ten children from disadvantaged social backgrounds displaying nonorganic behavioural disorders. The aim was to acquire novel ways of expression on the part of the participants to release painful emotions in a safe setting using the method of catharsis through acting and dancing, triggering relaxation as a physiological response and improving their attitude. This study retrospectively analyses the changes in the children’s behaviour and their active participation in the project through quantitative and qualitative research. The children improved their attention and self-esteem, their behaviour became less aggressive and impulsive, and half showed increased empathy. The active participation rate during the project was 82%.

Ahmed Al-Rawi

In this study, a mixed method was used to understand metajournalistic discourse on TikTok. Using convenience sampling, 30 journalists active on TikTok were selected and all their available 3270 TikTok videos were manually content analyzed. To understand these videos, a codebook was designed in which 5 major categories were identified as follows: 1. Educational news production; 2. News sharing/live reporting; 3. Personal anecdotes & opinions; 4. Interactions with audience; and 5. Dancing & challenges. In general, metajournalistic discourse on TikTok revolve around employing educational tips, entertainment, creativity, and music to engage and interact with wider audiences. To complement the study and gain a deeper understanding of journalism practice on TikTok, 17 journalists active on TikTok were interviewed mostly by email. Using thematic analysis, four major themes emerged from the interviews including personalized content, effective algorithms, audience outreach, and TikTok as a newsroom.

Tong Wang, Taotao Gu, Huan Deng et al.

As autonomous driving systems (ADS) advance towards higher levels of autonomy, orchestrating their safety verification becomes increasingly intricate. This paper unveils ScenarioFuzz, a pioneering scenario-based fuzz testing methodology. Designed like a choreographer who understands the past performances, it uncovers vulnerabilities in ADS without the crutch of predefined scenarios. Leveraging map road networks, such as OPENDRIVE, we extract essential data to form a foundational scenario seed corpus. This corpus, enriched with pertinent information, provides the necessary boundaries for fuzz testing in the absence of starting scenarios. Our approach integrates specialized mutators and mutation techniques, combined with a graph neural network model, to predict and filter out high-risk scenario seeds, optimizing the fuzzing process using historical test data. Compared to other methods, our approach reduces the time cost by an average of 60.3%, while the number of error scenarios discovered per unit of time increases by 103%. Furthermore, we propose a self-supervised collision trajectory clustering method, which aids in identifying and summarizing 54 high-risk scenario categories prone to inducing ADS faults. Our experiments have successfully uncovered 58 bugs across six tested systems, emphasizing the critical safety concerns of ADS.

Hsiang-Wei Huang, Cheng-Yen Yang, Wenhao Chai et al.

In the field of multi-object tracking (MOT), traditional methods often rely on the Kalman filter for motion prediction, leveraging its strengths in linear motion scenarios. However, the inherent limitations of these methods become evident when confronted with complex, nonlinear motions and occlusions prevalent in dynamic environments like sports and dance. This paper explores the possibilities of replacing the Kalman filter with a learning-based motion model that effectively enhances tracking accuracy and adaptability beyond the constraints of Kalman filter-based tracker. In this paper, our proposed method MambaMOT and MambaMOT+, demonstrate advanced performance on challenging MOT datasets such as DanceTrack and SportsMOT, showcasing their ability to handle intricate, non-linear motion patterns and frequent occlusions more effectively than traditional methods.

Alice Rixte

In Electronic Dance Music (EDM), many artists use DJing techniques in order to perform their own productions live. As a consequence, they do not have access during the performance to the internal structure of their tracks, and specifically to their equivalent of a partition: MIDI files. On the other hand, if an artist attempts to remix or interpret their own production live, the number of tracks that they can simultaneously control is limited without suitable software. This article introduces LiveScaler, a software that allows live control of the harmony and pitch of electronic music. A set of pitch transformations, termed affine transformations, is presented. These transformations are applied to all MIDI streams of a prepared track. A MaxMSP implementation, in conjunction with Ableton Live, is proposed. Special attention is given to control issues, mapping, and practical live experimentation in the context of EDM.

Linhong Chen, Xiaocang Xu

There are many studies on the impact of physical activity on health but few studies on the relationship between physical activity and medical expenditure among the elderly. Based on the China Health and Retirement Longitudinal Survey (CHARLS) database and selected 4456 valid samples, this paper used a two-part model to analyze the effects of high, moderate, and low physical activity intensity on medical expenditure. It is found that the intensity of physical activity was negatively correlated with medical expenditure, and the medical expenditure of the high physical activity intensity group was significantly lower than that of the low physical activity intensity group. For example, compared to people with no physical activity, the total medical expenditure decreased by 22.4%, 40.4%, and 62.5% per week in those with low, moderate, and high physical activity intensity. Thus, the government should provide more places for the elderly to exercise, planning special exercise areas for the elderly in community playgrounds, such as a dancing square, which will also help the elderly to increase their amount of exercise per week and develop a daily exercise habit.

Sunil K. Chebolu

This article is inspired by the great conjunction of Jupiter and Saturn on December 21st, 2020. We will address the following questions from a mathematical standpoint. What is a conjunction? How often do we have a conjunction of two planets? How are great conjunctions distributed in the sky? How long will it take for a cycle of great conjunctions to return to the same point in the sky and time of the year? Can there be a conjunction of 3 outer planets? How often can that happen? We will begin by setting up the geometric framework for formulating these questions. Then we will analyze these questions by first examining similar questions in a more mundane context: the three hands in a clock. Even though the clock model is a simple model that ignores the more subtle and complex elements of planets' cosmic dance, it captures the main ideas succinctly.