arXiv subjects

Jisoo Park

Publications and source records attributed to Jisoo Park.

5 recordsLinked to original sources

arXiv2026

URHead: A Unified UV-Space Representation for Joint Mesh-3DGS Optimization in Head Avatars

We present URHead, a unified representation for high-fidelity and animatable head avatars that fundamentally redefines mesh-Gaussian integration. While mesh-based methods offer precise geometric control but lack photorealistic detail, and Gaussian-based approaches achieve photorealism but suffer from poor structural consistency, existing hybrid solutions fail to fully leverage their complementary strengths. Our key contribution is a UV-space unification where both representations share a common UV parameterization. Through joint optimization with adaptive gaussian sampling, our method automatically learns to disentangle and allocate appropriate roles to each component. URHead maintains full parametric controllability while preserving subject-specific details, and outperforms existing state-of-the-art methods in reconstruction quality and animation consistency.

cs.GR↗

arXiv2026

$C^3$ASD: Multi-Level Consistency-Driven Representation Learning

Active Speaker Detection determines whether a visible person in a video is speaking at each moment. While recent audio-visual fusion methods perform well on clean data, they degrade under real-world corruptions such as background noise, occlusion, or simultaneous modality degradation. We attribute this limitation to the absence of explicit consistency constraints that promote robust, semantically aligned representations across modalities. Without such guidance, models tend to learn fragile modality-specific shortcuts that fail under corrupted conditions. We propose $C^3$ASD, a multi-level consistency-driven framework with three complementary constraints: embedding-level inter-modality consistency aligns audio-visual representations during speech; sequence-level intra-modality consistency separates speaking and non-speaking clusters via track-aware contrastive learning; and prediction-level consistency stabilizes fusion through knowledge distillation. Extensive experiments demonstrate significant improvements under diverse audio, visual and joint corruptions, while maintaining competitive performance on clean data.

cs.CV↗

arXiv2026

Modeling conflicting incentives in engineering senior capstone projects: A multi-player game theory approach

University engineering capstone projects involve sustained interaction among students, faculty, and industry sponsors whose objectives are only partially aligned. While capstones are widely used in engineering education, existing analyses typically treat stakeholder behavior informally or descriptively, leaving incentive conflicts, information asymmetries, and strategic dependencies underexplored. This paper develops a formal game-theoretic framework that models capstone projects as a sequential Bayesian game involving three players: the university, the industry sponsor, and the student team. The framework is intended as an analytical and explanatory tool for understanding how institutional policy choices, such as grading structures, intellectual property rules, and sponsor engagement expectations, shape stakeholder behavior and project outcomes, rather than as a calibrated or predictive model. The university acts as a constrained Stackelberg leader by committing to course policies and assessment structures while anticipating strategic responses by sponsors and students under incomplete information. Reduced-form outcome functions capture technical quality, documentation quality, timeliness, alignment with sponsor needs, and publishability, while payoff functions reflect stakeholder-specific objectives and costs. Under standard assumptions, the model admits stable equilibrium regimes that correspond to empirically recognizable capstone dynamics observed in practice, including cooperative engagement, sponsor-dominated exploitation, and student grade gaming. Rather than claiming precise prediction, the framework provides a structured basis for reasoning about incentive design, policy tradeoffs, and structural failure modes in project-based learning environments, as well as for future extensions incorporating richer dynamics, repeated interaction, and empirical calibration.

cs.CY↗

arXiv2025

Lightweight Wasserstein Audio-Visual Model for Unified Speech Enhancement and Separation

Speech Enhancement (SE) and Speech Separation (SS) have traditionally been treated as distinct tasks in speech processing. However, real-world audio often involves both background noise and overlapping speakers, motivating the need for a unified solution. While recent approaches have attempted to integrate SE and SS within multi-stage architectures, these approaches typically involve complex, parameter-heavy models and rely on supervised training, limiting scalability and generalization. In this work, we propose UniVoiceLite, a lightweight and unsupervised audio-visual framework that unifies SE and SS within a single model. UniVoiceLite leverages lip motion and facial identity cues to guide speech extraction and employs Wasserstein distance regularization to stabilize the latent space without requiring paired noisy-clean data. Experimental results demonstrate that UniVoiceLite achieves strong performance in both noisy and multi-speaker scenarios, combining efficiency with robust generalization. The source code is available at https://github.com/jisoo-o/UniVoiceLite.

cs.CV↗

arXiv2023

A Hybrid Semantic-Geometric Approach for Clutter-Resistant Floorplan Generation from Building Point Clouds

Building Information Modeling (BIM) technology is a key component of modern construction engineering and project management workflows. As-is BIM models that represent the spatial reality of a project site can offer crucial information to stakeholders for construction progress monitoring, error checking, and building maintenance purposes. Geometric methods for automatically converting raw scan data into BIM models (Scan-to-BIM) often fail to make use of higher-level semantic information in the data. Whereas, semantic segmentation methods only output labels at the point level without creating object level models that is necessary for BIM. To address these issues, this research proposes a hybrid semantic-geometric approach for clutter-resistant floorplan generation from laser-scanned building point clouds. The input point clouds are first pre-processed by normalizing the coordinate system and removing outliers. Then, a semantic segmentation network based on PointNet++ is used to label each point as ceiling, floor, wall, door, stair, and clutter. The clutter points are removed whereas the wall, door, and stair points are used for 2D floorplan generation. A region-growing segmentation algorithm paired with geometric reasoning rules is applied to group the points together into individual building elements. Finally, a 2-fold Random Sample Consensus (RANSAC) algorithm is applied to parameterize the building elements into 2D lines which are used to create the output floorplan. The proposed method is evaluated using the metrics of precision, recall, Intersection-over-Union (IOU), Betti error, and warping error.

cs.CV↗