DSAR: Dual-Stream Autoregressive Modeling of Temporal Cloth Dynamics for Photorealistic Animatable Avatars

• URL: http://arxiv.org/abs/2608.10500v1
• Date: Tue, 11 Aug 2026 05:19:21 GMT
• Title: DSAR: Dual-Stream Autoregressive Modeling of Temporal Cloth Dynamics for Photorealistic Animatable Avatars
• Abstract summary: We introduce a dual-stream autoregressive framework that explicitly models both observable geometric information and implicit internal state.<n> Experiments on challenging datasets demonstrate significant improvements in rendering quality, temporal consistency, and generalization to motion patterns beyond training distributions.
• Score: 7.198815179389743
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Creating photorealistic and temporally coherent animatable human avatars from RGB videos remains challenging. Current methods struggle to capture realistic cloth dynamics, producing over-smoothed appearance or severe artifacts on out-of-distribution poses. This limitation stems from a fundamental oversight: existing approaches neglect the temporal causality inherent in cloth physics, where current states emerge from previous states through temporal evolution rather than instantaneous skeletal configurations alone. Without explicit modeling of this causal structure, networks learn pose-appearance correlations instead of motion evolution, leading to poor generalization. We introduce a dual-stream autoregressive framework that explicitly models both observable geometric information and implicit internal state. The geometric stream propagates surface displacement from the previous frame, while the state stream fuses current features with historical states retrieved from a memory bank. Motion-adaptive aggregation handles spatially-varying dynamics, and adaptive regularization balances smoothness with flexibility. Experiments on challenging datasets demonstrate significant improvements in rendering quality, temporal consistency, and generalization to motion patterns beyond training distributions, validating that dual-stream temporal modeling enables realistic cloth dynamics.

Related papers

• DyG$^2$T: Modeling Object Dynamics with 3D Gaussian Temporal-Spatial Particle Graph Transformer [86.4618122245946]
  DyG$2$T is a dynamics modeling framework that infers object motion trajectories by spatially completing and temporally discriminating Key Point representations.<n>Experiments on both synthetic and real-world datasets demonstrate that DyG$2$T achieves accurate dynamics modeling and reasoning.
  arXiv  Detail & Related papers  (2026-08-19T03:44:28Z)
• DynaOD: Dynamic Origin-Destination Flow Generation with Discrete-to-Continuous Temporal Semantic Modeling [20.859733498481173]
  Dynamic origin-destination (OD) flow generation seeks to synthesize realistic mobility dynamics from temporal context alone.<n>We propose DynaOD, a semantic-driven framework that models temporal dynamics through two complementary perspectives.<n>Our method consistently outperforms representative baselines in both predictive accuracy and distributional fidelity.
  arXiv  Detail & Related papers  (2026-06-08T06:32:49Z)
• MORPHOS: Autoregressive 4D Generation with Temporal Structured Latents [35.76764167284739]
  We present MORPHOS, a novel autoregressive framework that generates dynamic 3D assets from videos across diverse representations.<n>We introduce the Temporal Structured Latents (T-SLAT), a unified 4D representation that jointly encodes geometry and appearance along the temporal dimension.
  arXiv  Detail & Related papers  (2026-06-01T17:01:21Z)
• Latent Dynamics for Full Body Avatar Animation [18.84485573885924]
  We augment a pose-conditioned 3D Gaussian avatar with a transformer-based decoder and a dynamics residual latent.<n>The model decomposes each update into driving, restoring, and dissipative forces, producing temporally coherent, history-dependent rollouts.<n>Across nine captured sequences of everyday motion with diverse loose garments, quantitative metrics and a perceptual user study show improved animation quality over recent data-driven baselines.
  arXiv  Detail & Related papers  (2026-05-20T17:58:03Z)
• Lie Flow: Video Dynamic Fields Modeling and Predicting with Lie Algebra as Geometric Physics Principle [48.28007238304401]
  LieFlow is a dynamic radiance representation framework that explicitly models motion.<n>The SE(3) transformation field enforces physically inspired constraints to maintain motion continuity and geometric consistency.<n>Results confirm that SE(3)-based motion modeling offers a robust and physically grounded framework for representing dynamic 4D scenes.
  arXiv  Detail & Related papers  (2026-02-25T07:19:18Z)
• SplitGaussian: Reconstructing Dynamic Scenes via Visual Geometry Decomposition [14.381223353489062]
  We propose textbfSplitGaussian, a novel framework that explicitly decomposes scene representations into static and dynamic components.<n>SplitGaussian outperforms prior state-of-the-art methods in rendering quality, geometric stability, and motion separation.
  arXiv  Detail & Related papers  (2025-08-06T09:00:13Z)
• VDEGaussian: Video Diffusion Enhanced 4D Gaussian Splatting for Dynamic Urban Scenes Modeling [68.65587507038539]
  We present a novel video diffusion-enhanced 4D Gaussian Splatting framework for dynamic urban scene modeling.<n>Our key insight is to distill robust, temporally consistent priors from a test-time adapted video diffusion model.<n>Our method significantly enhances dynamic modeling, especially for fast-moving objects, achieving an approximate PSNR gain of 2 dB.
  arXiv  Detail & Related papers  (2025-08-04T07:24:05Z)
• FlowMo: Variance-Based Flow Guidance for Coherent Motion in Video Generation [51.110607281391154]
  FlowMo is a training-free guidance method for enhancing motion coherence in text-to-video models.<n>It estimates motion coherence by measuring the patch-wise variance across the temporal dimension and guides the model to reduce this variance dynamically during sampling.
  arXiv  Detail & Related papers  (2025-06-01T19:55:33Z)
• Dynamical Diffusion: Learning Temporal Dynamics with Diffusion Models [71.63194926457119]
  We introduce Dynamical Diffusion (DyDiff), a theoretically sound framework that incorporates temporally aware forward and reverse processes.<n>Experiments across scientifictemporal forecasting, video prediction, and time series forecasting demonstrate that Dynamical Diffusion consistently improves performance in temporal predictive tasks.
  arXiv  Detail & Related papers  (2025-03-02T16:10:32Z)
• Real-time Deep Dynamic Characters [95.5592405831368]
  We propose a deep videorealistic 3D human character model displaying highly realistic shape, motion, and dynamic appearance. We use a novel graph convolutional network architecture to enable motion-dependent deformation learning of body and clothing. We show that our model creates motion-dependent surface deformations, physically plausible dynamic clothing deformations, as well as video-realistic surface textures at a much higher level of detail than previous state of the art approaches.
  arXiv  Detail & Related papers  (2021-05-04T23:28:55Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.