arXiv subjects

Ding Fan

Publications and source records attributed to Ding Fan.

3 recordsLinked to original sources

arXiv2026

Who Responds When the Driver Is Gone? A Framework for Holistic Passenger Intent Understanding

As autonomous vehicles advance toward driverless mobility, understanding and responding to passenger needs and intentions becomes increasingly important in the absence of a human driver. We propose Intent2Drive, a unified framework for holistic passenger intent understanding and passenger-aligned planning. Unlike existing methods that rely on explicit commands, Intent2Drive models passenger intent as a latent cognitive state inferred from language, personal attributes, emotions, behaviors, and situational context. To support this task, we construct the Holistic Passenger Intent Dataset (HPID) with structured annotations of explicit and implicit passenger-intent cues. A Theory-of-Mind-inspired Passenger Intent Reasoner (PIR) infers a Latent Passenger State (LPS) and converts it into a planner-compatible Passenger Intent Objective (PIO). We validate the downstream utility of PIO by conditioning an existing hierarchical planning pipeline at the route and trajectory levels. Experiments demonstrate that the proposed method understands and responds to passenger needs, enabling passenger-aligned driving while maintaining competitive closed-loop planning performance.

cs.HC↗

arXiv2026

Functional MRI Time Series Generation via Wavelet-Based Image Transform and Spectral Flow Matching for Brain Disorder Identification

Functional Magnetic Resonance Imaging (fMRI) provides non-invasive access to dynamic brain activity by measuring blood oxygen level-dependent (BOLD) signals over time. However, the resource-intensive nature of fMRI acquisition limits the availability of high-fidelity samples required for data-driven brain analysis models. While modern generative models can synthesize fMRI data, they often remain challenging in replicating their inherent non-stationarity, intricate spatiotemporal dynamics, and physiological variations of raw BOLD signals. To address these challenges, we propose Dual-Spectral Flow Matching (DSFM), a novel fMRI generative framework that cascades dual frequency representation of BOLD signals with spectral flow matching. Specifically, our framework first converts BOLD signals into a wavelet decomposition map via a discrete wavelet transform (DWT) to capture globalized transient and multi-scale variations, and projects into the discrete cosine transform (DCT) space across brain regions and time to exploit localized energy compaction of low-frequency dominant BOLD coefficients. Subsequently, a spectral flow matching model is trained to generate class-conditioned cosine-frequency representation. The generated samples are reconstructed through inverse DCT and inverse DWT operations to recover physiologically plausible time-domain BOLD signals. This dual-transform approach imposes structured frequency priors and preserves key physiological brain dynamics. Ultimately, we demonstrate the efficacy of our approach through improved downstream fMRI-based brain network classification. The code is available at https://github.com/htew0001/DSFM.git .

cs.LG↗

arXiv2025

Physics-Informed Deformable Gaussian Splatting: Towards Unified Constitutive Laws for Time-Evolving Material Field

Recently, 3D Gaussian Splatting (3DGS), an explicit scene representation technique, has shown significant promise for dynamic novel-view synthesis from monocular video input. However, purely data-driven 3DGS often struggles to capture the diverse physics-driven motion patterns in dynamic scenes. To fill this gap, we propose Physics-Informed Deformable Gaussian Splatting (PIDG), which treats each Gaussian particle as a Lagrangian material point with time-varying constitutive parameters and is supervised by 2D optical flow via motion projection. Specifically, we adopt static-dynamic decoupled 4D decomposed hash encoding to reconstruct geometry and motion efficiently. Subsequently, we impose the Cauchy momentum residual as a physics constraint, enabling independent prediction of each particle's velocity and constitutive stress via a time-evolving material field. Finally, we further supervise data fitting by matching Lagrangian particle flow to camera-compensated optical flow, which accelerates convergence and improves generalization. Experiments on a custom physics-driven dataset as well as on standard synthetic and real-world datasets demonstrate significant gains in physical consistency and monocular dynamic reconstruction quality.

cs.CV↗