arXiv subjects

Hideaki Uchiyama

Publications and source records attributed to Hideaki Uchiyama.

12 recordsLinked to original sources

arXiv2026

ARMS: Anchor-Relational Motion Streaming for Seamless Solo-Social Motion Transitions

Generating temporally continuous and socially coherent human motion from text remains a fundamental challenge, particularly in realistic streams where people act alone, enter interactions, and later disengage. Most existing methods generate fixed-length motion clips under static agent configurations, which makes them brittle to solo-social transitions and unsuitable for incremental generation over long horizons. We propose ARMS, an Anchor-Relational Motion Streaming framework that unifies solo motion and human-human interaction within a single causal generative process. ARMS introduces a dynamics-asymmetric representation that decouples per-person temporal evolution from inter-person alignment via a partner-referenced relative-translation term, enabling seamless switching of social coupling without sacrificing long-horizon stability or spatial consistency between agents. On top of a causal latent space, a causal relational diffusion model progressively refines motion segment by segment using only past context, capturing both intra-person temporal dependencies and inter-person relations. Mode-aware relational gating activates or masks cross-agent connections, allowing the same model to support both solo and interaction generation. Experiments show that ARMS improves transition smoothness and social coherence compared to interaction-centric baselines, while also achieving competitive results on human-human interaction benchmarks.

cs.CV↗

arXiv2026

Generative Learning as a Tool to Improve Perception of Emotional Body Motion Expressions

Emotional body motion expressions are an essential element of non-verbal communication. Effectively conveying these expressions through technology is of utmost importance, for example, with virtual reality avatars and in social robotics. Recent advances in generative models have opened new opportunities for advancing research on emotional body motion learning. However, generating accurate emotional expression representations is challenging, given the subtlety of emotional cues, individual variability, and cultural differences. We investigate whether a generative model can implicitly learn emotional body motions directly from culturally grounded motion-capture data, without explicit emotion-motion guidance. Using a dataset of emotional performances by 49 Japanese actors, we trained a Transformer-based generative model to generate expressive motions conditioned on 13 discrete emotion labels. We evaluate the generated motions from two perspectives: (1) an LSTM-based classifier to assess recognizability by machine observers, achieving a recognition accuracy of 22.80%, and (2) a human perception study with Japanese raters to assess alignment with human affective interpretations, yielding a recognition accuracy of 24.91%. Beyond these, we evaluate the utility of generative modeling for three practical tasks: augmenting emotion recognition models, extracting representative emotion-specific motion patterns, and synthesizing smooth transitions between emotion intensities. Our findings highlight the potential of implicit, data-driven generative modeling to enhance affective computing applications and our understanding of emotion expressions.

cs.LG↗

arXiv2026

EMA: Effort Metric Attention for Anatomical Effort-Guided Human Motion Diffusion

Human motion diffusion models can synthesize action sequences from text, but controlling motion intensity remains challenging. Existing approaches rely on effort-related adverbs, which are ambiguous and fail to capture quantitative aspects such as pacing, often resulting in flat and monotonous dynamics. We propose an intensity-control framework based on Effort Metric Attention (EMA), a cross-attention module that conditions diffusion on numerical effort signals. Inspired by Laban Movement Analysis (LMA), the framework focuses on the Time and Weight effort factors. We approximate these factors using two kinematic metrics: peak joint positional change for pacing and collective joint positional change for motion amount. EMA enables fine-grained, region-wise control without costly post-hoc optimization. We introduce two evaluation tasks, metric-to-motion consistency and body-part-level effort modulation, to assess numerical fidelity and localized control. Experiments and a user study show near-monotonic alignment between specified effort levels, generated motion dynamics, and established LMA descriptors. These results indicate effective and interpretable control of effort dynamics in practice.

cs.CV↗

arXiv2025

Mind Your Vision: Multimodal Estimation of Refractive Disorders Using Electrooculography and Eye Tracking

Refractive errors are among the most common visual impairments globally, yet their diagnosis often relies on active user participation and clinical oversight. This study explores a passive method for estimating refractive power using two eye movement recording techniques: electrooculography (EOG) and video-based eye tracking. Using a publicly available dataset recorded under varying diopter conditions, we trained Long Short-Term Memory (LSTM) models to classify refractive power from unimodal (EOG or eye tracking) and multimodal configuration. We assess performance in both subject-dependent and subject-independent settings to evaluate model personalization and generalizability across individuals. Results show that the multimodal model consistently outperforms unimodal models, achieving the highest average accuracy in both settings: 96.207\% in the subject-dependent scenario and 8.882\% in the subject-independent scenario. However, generalization remains limited, with classification accuracy only marginally above chance in the subject-independent evaluations. Statistical comparisons in the subject-dependent setting confirmed that the multimodal model significantly outperformed the EOG and eye-tracking models. However, no statistically significant differences were found in the subject-independent setting. Our findings demonstrate both the potential and current limitations of eye movement data-based refractive error estimation, contributing to the development of continuous, non-invasive screening methods using EOG signals and eye-tracking data.

eess.IV↗

arXiv2025

UMotion: Uncertainty-driven Human Motion Estimation from Inertial and Ultra-wideband Units

Sparse wearable inertial measurement units (IMUs) have gained popularity for estimating 3D human motion. However, challenges such as pose ambiguity, data drift, and limited adaptability to diverse bodies persist. To address these issues, we propose UMotion, an uncertainty-driven, online fusing-all state estimation framework for 3D human shape and pose estimation, supported by six integrated, body-worn ultra-wideband (UWB) distance sensors with IMUs. UWB sensors measure inter-node distances to infer spatial relationships, aiding in resolving pose ambiguities and body shape variations when combined with anthropometric data. Unfortunately, IMUs are prone to drift, and UWB sensors are affected by body occlusions. Consequently, we develop a tightly coupled Unscented Kalman Filter (UKF) framework that fuses uncertainties from sensor data and estimated human motion based on individual body shape. The UKF iteratively refines IMU and UWB measurements by aligning them with uncertain human motion constraints in real-time, producing optimal estimates for each. Experiments on both synthetic and real-world datasets demonstrate the effectiveness of UMotion in stabilizing sensor data and the improvement over state of the art in pose accuracy.

cs.GR↗

arXiv2025

MagicCraft: Natural Language-Driven Generation of Dynamic and Interactive 3D Objects for Commercial Metaverse Platforms

Metaverse platforms are rapidly evolving to provide immersive spaces for user interaction and content creation. However, the generation of dynamic and interactive 3D objects remains challenging due to the need for advanced 3D modeling and programming skills. To address this challenge, we present MagicCraft, a system that generates functional 3D objects from natural language prompts for metaverse platforms. MagicCraft uses generative AI models to manage the entire content creation pipeline: converting user text descriptions into images, transforming images into 3D models, predicting object behavior, and assigning necessary attributes and scripts. It also provides an interactive interface for users to refine generated objects by adjusting features such as orientation, scale, seating positions, and grip points. Implemented on Cluster, a commercial metaverse platform, MagicCraft was evaluated by 7 expert CG designers and 51 general users. Results show that MagicCraft significantly reduces the time and skill required to create 3D objects. Users with no prior experience in 3D modeling or programming successfully created complex, interactive objects and deployed them in the metaverse. Expert feedback highlighted the system's potential to improve content creation workflows and support rapid prototyping. By integrating AI-generated content into metaverse platforms, MagicCraft makes 3D content creation more accessible.

cs.HC↗

arXiv2024

MagicItem: Dynamic Behavior Design of Virtual Objects with Large Language Models in a Consumer Metaverse Platform

To create rich experiences in virtual reality (VR) environments, it is essential to define the behavior of virtual objects through programming. However, programming in 3D spaces requires a wide range of background knowledge and programming skills. Although Large Language Models (LLMs) have provided programming support, they are still primarily aimed at programmers. In metaverse platforms, where many users inhabit VR spaces, most users are unfamiliar with programming, making it difficult for them to modify the behavior of objects in the VR environment easily. Existing LLM-based script generation methods for VR spaces require multiple lengthy iterations to implement the desired behaviors and are difficult to integrate into the operation of metaverse platforms. To address this issue, we propose a tool that generates behaviors for objects in VR spaces from natural language within Cluster, a metaverse platform with a large user base. By integrating LLMs with the Cluster Script provided by this platform, we enable users with limited programming experience to define object behaviors within the platform freely. We have also integrated our tool into a commercial metaverse platform and are conducting online experiments with 63 general users of the platform. The experiments show that even users with no programming background can successfully generate behaviors for objects in VR spaces, resulting in a highly satisfying system. Our research contributes to democratizing VR content creation by enabling non-programmers to design dynamic behaviors for virtual objects in metaverse platforms.

cs.HC↗

arXiv2023

Ensemble Learning to Assess Dynamics of Affective Experience Ratings and Physiological Change

The congruence between affective experiences and physiological changes has been a debated topic for centuries. Recent technological advances in measurement and data analysis provide hope to solve this epic challenge. Open science and open data practices, together with data analysis challenges open to the academic community, are also promising tools for solving this problem. In this entry to the Emotion Physiology and Experience Collaboration (EPiC) challenge, we propose a data analysis solution that combines theoretical assumptions with data-driven methodologies. We used feature engineering and ensemble selection. Each predictor was trained on subsets of the training data that would maximize the information available for training. Late fusion was used with an averaging step. We chose to average considering a ``wisdom of crowds'' strategy. This strategy yielded an overall RMSE of 1.19 in the test set. Future work should carefully explore if our assumptions are correct and the potential of weighted fusion.

cs.LG↗

arXiv2023

Smart Dimming Sunglasses for Photophobia Using Spatial Light Modulator

We present a smart sunglasses system engineered to assist individuals experiencing photophobia, particularly those highly sensitive to light intensity. The system integrates a high dynamic range (HDR) camera and a liquid crystal spatial light modulator (SLM) to dynamically regulate light, adapting to environmental scenes by modifying pixel transmittance through a specialized control algorithm, thereby offering adaptable light management to meet the users' visual needs. Nonetheless, a conventional occlusion mask on the SLM, intended to block incoming light, emerges blurred and insufficient due to a misaligned focal plane. To address the challenge of imprecise light filtering, we introduce an optimization algorithm that meticulously adjusts the light attenuation process, effectively diminishing excessive brightness in targeted areas without adversely impacting regions with acceptable levels of luminance.

cs.HC↗

arXiv2023

Toward unlabeled multi-view 3D pedestrian detection by generalizable AI: techniques and performance analysis

We unveil how generalizable AI can be used to improve multi-view 3D pedestrian detection in unlabeled target scenes. One way to increase generalization to new scenes is to automatically label target data, which can then be used for training a detector model. In this context, we investigate two approaches for automatically labeling target data: pseudo-labeling using a supervised detector and automatic labeling using an untrained detector (that can be applied out of the box without any training). We adopt a training framework for optimizing detector models using automatic labeling procedures. This framework encompasses different training sets/modes and multi-round automatic labeling strategies. We conduct our analyses on the publicly-available WILDTRACK and MultiviewX datasets. We show that, by using the automatic labeling approach based on an untrained detector, we can obtain superior results than directly using the untrained detector or a detector trained with an existing labeled source dataset. It achieved a MODA about 4% and 1% better than the best existing unlabeled method when using WILDTRACK and MultiviewX as target datasets, respectively.

cs.CV↗

arXiv2022

MOTSLAM: MOT-assisted monocular dynamic SLAM using single-view depth estimation

Visual SLAM systems targeting static scenes have been developed with satisfactory accuracy and robustness. Dynamic 3D object tracking has then become a significant capability in visual SLAM with the requirement of understanding dynamic surroundings in various scenarios including autonomous driving, augmented and virtual reality. However, performing dynamic SLAM solely with monocular images remains a challenging problem due to the difficulty of associating dynamic features and estimating their positions. In this paper, we present MOTSLAM, a dynamic visual SLAM system with the monocular configuration that tracks both poses and bounding boxes of dynamic objects. MOTSLAM first performs multiple object tracking (MOT) with associated both 2D and 3D bounding box detection to create initial 3D objects. Then, neural-network-based monocular depth estimation is applied to fetch the depth of dynamic features. Finally, camera poses, object poses, and both static, as well as dynamic map points, are jointly optimized using a novel bundle adjustment. Our experiments on the KITTI dataset demonstrate that our system has reached best performance on both camera ego-motion and object tracking on monocular dynamic SLAM.

cs.CV↗

arXiv2020

TetraTSDF: 3D human reconstruction from a single image with a tetrahedral outer shell

Recovering the 3D shape of a person from its 2D appearance is ill-posed due to ambiguities. Nevertheless, with the help of convolutional neural networks (CNN) and prior knowledge on the 3D human body, it is possible to overcome such ambiguities to recover detailed 3D shapes of human bodies from single images. Current solutions, however, fail to reconstruct all the details of a person wearing loose clothes. This is because of either (a) huge memory requirement that cannot be maintained even on modern GPUs or (b) the compact 3D representation that cannot encode all the details. In this paper, we propose the tetrahedral outer shell volumetric truncated signed distance function (TetraTSDF) model for the human body, and its corresponding part connection network (PCN) for 3D human body shape regression. Our proposed model is compact, dense, accurate, and yet well suited for CNN-based regression task. Our proposed PCN allows us to learn the distribution of the TSDF in the tetrahedral volume from a single image in an end-to-end manner. Results show that our proposed method allows to reconstruct detailed shapes of humans wearing loose clothes from single RGB images.

cs.CV↗