The results show that the proposed SkelGen4D weakly supervised skeleton modeling matches or surpasses fully supervised baselines while scaling to diverse object categories for high-quality text-driven mesh animation.
Abstract
We study 4D generation to synthesize temporally coherent sequences of 3D geometry for animation and content creation. In contrast to existing SDS-based optimization methods and video-driven animation approaches, we adopt a skeleton-driven animation framework aligned with standard industrial pipelines, which enables explicit control and editing. To this end, we propose SkelGen4D, a weakly supervised feed-forward framework for text-driven mesh animation that generates explicit skeleton motions without requiring per-frame skeleton annotations. SkelGen4D first recovers temporally consistent pseudo-skeletons from animated meshes via differentiable fitting, and then generates text-conditioned skeleton motion sequences in a feed-forward manner, further refined with Motion-GRPO to ensure temporally coherent, physically plausible, and articulated animation. We evaluate our method on two large-scale benchmarks, Truebones Zoo and Diffusion4D. Our results show that our weakly supervised skeleton modeling matches or surpasses fully supervised baselines while scaling to diverse object categories for high-quality text-driven mesh animation. Further, our method supports flexible motion editing and is aligned with standard animation production pipelines.
Recent advances in 4D content generation have attracted increasing attention, yet creating high-quality animated 3D models remains challenging due to the complexity of modeling spatio-temporal distributions and the scarcity of 4D training data. We present AnimateAnyMesh++, a feed-forward framework for text-driven animation of arbitrary 3D meshes with substantial upgrades in data, architecture, and generative capability. First, we expand the DyMesh-XL dataset by mining dynamic content from Objaverse-XL, increasing the number of unique identities from 60K to 300K and substantially broadening category and motion diversity. Second, we redesign DyMeshVAE-Flex with power-law topology-aware attention and vertex-normal-enhanced features, which significantly improves trajectory reconstruction, local geometry preservation, and mit igates trajectory-sticking artifacts. Third, we introduce archi tectural changes to both DyMeshVAE-Flex and the rectified flow (RF) generator to support variable-length sequence training and generation, enabling longer animations while preserving reconstruction fidelity. Extensive experiments demonstrate that AnimateAnyMesh++ generates semantically accurate and tem porally coherent mesh animations within seconds, surpassing prior approaches in quality and efficiency. The enlarged DyMesh XL, the upgraded DyMeshVAE-Flex, and variable-length RF to gether deliver consistent gains across benchmarks and in-the-wild meshes. We will release code, models, and the expanded DyMesh XL at https://github.com/JarrentWu1031/AnimateAnyMesh-pp upon acceptance of this manuscript to facilitate research in 4D content creation.
Zijie Wu, Chaohui Yu, Fan Wang et al.· IEEE Transactions on Pattern...· 2 citations· ⚡1
ArtiMo, a novel agent-driven framework for text-guided articulated mesh animation, develops an agentic pipeline powered by Large Language and Vision-Language Models (LLMs/VLMs) to orchestrate motion generation without requiring model fine-tuning.
Chunyu Zou, Peng Dai, Yi-Hua Huang et al.· 0 citations
Recent advances in automatic rigging now deliver animation-ready 3D assets at scale, yet generating the motion to drive them remains a bottleneck. Existing learned animators are topology-constrained: they rely on category-specific templates or require per-skeleton fine-tuning and reference motions at inference. We present UniMate, a unified foundation model that synthesizes articulated motion for arbitrary skeletons from a rigged 3D asset and a text prompt, with no test-time optimization or per-skeleton retraining. UniMate introduces a topology-aware diffusion transformer, which integrates skeletal topology into attention via three mechanisms: (1) a graph-aware attention bias from pairwise joint relations and geodesic distances; (2) a spectral rotary position embedding generalizing RoPE to arbitrary kinematic trees via the graph Laplacian; and (3) a global topological conditioner attention-pooled from the rest-pose skeleton. We also curate UniML3D, 13,006 motion sequences spanning bipedal, quadrupedal, avian, marine, insectoid, serpentine, and articulated rigid objects with unified canonicalization and text pairing. Trained on this dataset, UniMate outperforms state-of-the-art baselines in quality, generalization, and efficiency, and supports zero-shot cross-topology transfer, in-betweening, expansion, and text-guided editing. Our project page is available at https://linzhanmou.com/unimate/.
Lin-Zhan Mou, Jiahui Lei, Zhi-Yang Dou et al.· 0 citations
We introduce BLARM, a feed-forward method for video-driven 3D mesh animation. Given a monocular video and a static object mesh, BLARM predicts a temporally coherent animated mesh whose motion follows the video. Rather than relying on explicit rigs or directly regressing high-dimensional vertex motion, we represent animation using a compact set of learned, time-varying rigid motion components and time-invariant vertex-to-component skinning weights. This yields a low-dimensional deformation space without requiring skeletons, cages, skinning weights, or rig annotations. Our architecture conditions geometry-derived deformation latents on video features through factorized spatial-temporal attention, then decodes rigid transformations blended by predicted skinning weights. Trained with trajectory reconstruction, entropy regularization, and motion-aware contrastive learning, BLARM produces accurate and temporally stable animations while recovering compact, interpretable motion structure from monocular video.
Pradyumn Goyal, Yizhak Ben-Shabat, Hsueh-Ti Derek Liu et al.· 0 citations
The convergence of diffusion models and 3D Gaussian Splatting (3D -GS) has catalyzed a paradigm shift in AI-driven 3D/4D content creation, enabling high-quality generation, editing, and animation from natural language instructions. This narrative review syn thesizes representative works published between 2023 and 2025 across four interconnected sub-areas: text-to-3D/4D generation, 3D scene editing, dynamic scene animation, and motion and video generation. Through comparative analysis of representation types, supervision strategies, and temporal modeling approaches, we identify convergent design principles—coarse- to-fine optimization in distillation-based methods, explicit representations, regularization against score - distillation variance, and decomposition for controllability—that recur across sub-areas. We catalog six critical limitations and propose concrete directions for improvement for each. This review serves as a foundational reference for researchers entering this rapidly evolving field.
XR content creation increasingly combines reconstruction, generative modeling, animation synthesis, neural rendering, and real-time engine deployment. However, creating deployable synthetic 4D humans remains fragmented because human assets must preserve body structure, appearance, motion, deformation, and scene-level plausibility across time. This paper proposes SPL4SH, a Systematic Pipeline for 4D Synthetic Humans, to investigate how contemporary AI-assisted techniques can be organized into an XR-oriented production workflow. SPL4SH is built from a stage-based taxonomy covering modeling, rigging, animation, rendering, and interaction. This taxonomy supports a comparative suitability assessment of recent techniques and guides a modular proof-of-concept implementation integrating SMPLify for parametric body modeling and rigging, Kimodo for controllable motion generation, SMPLitex for SMPL-compatible texture generation, Rokoko-based retargeting, Blender-based asset integration, and Unreal Engine deployment. The evaluation combines individual module tests, full-pipeline integration, and scene-level deployment in a meeting-room environment involving human-object, human-human, and human-scene arrangements. Results show that modular 4D human generation is technically feasible: SMPL-X geometry supports rigged deformation, generated motions can be retargeted to animated characters, texture maps improve mannequin-like bodies, and final assets can be placed in real-time XR scenes. However, persistent barriers remain, including format mismatch, manual skeleton alignment, retargeting fragility, texture discontinuities, hallucinated visual regions, motion interpenetration, stiff transitions, and scene-level validation. SPL4SH contributes a grounded framework for selecting, combining, and evaluating independent generative components as XR-ready synthetic human assets.
David Ohara de Souza Cardoso, Willams de Lima Costa, Pedro Azevedo Abrantes de Oliveira et al.· Journal of the Brazilian Com...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
