Searcharxiv⌕ Search

arXiv subjects

Zijiao Zeng

Publications and source records attributed to Zijiao Zeng.

12 recordsLinked to original sources

MoRAE: Flow-Friendly Self-Supervised Latents for Text-to-Motion Generation

Which latent space should a text-to-motion generator model? We show that a self-supervised motion representation, once compressed, is a better generative space than latents learned by reconstruction. Our approach, MoRAE, is simple: we pre-train a Motion-JEPA encoder that predicts hidden joint trajectories from the rest of the body, and compress its frozen features into a compact latent for a standard flow transformer. Two observations drive this design. First, prediction yields features that capture action semantics and cross-joint structure better than reconstruction. Second, these features are ill-conditioned, and a standard flow transformer generates poorly in them directly. Compressing them into a compact latent makes generation reliable and places cross-joint structure on the directions the flow models most accurately. With a 58M DiT and the standard CLIP text condition, MoRAE achieves state-of-the-art FID on HumanML3D and KIT-ML while maintaining retrieval-based text consistency and diversity close to real motion.

cs.CV↗

FlashMotion: Ultra-Fast Long-Term 3D Motion Generation

3D human motion generation plays a crucial role in game development, extended reality (XR), digital humans, and embodied simulation. Beyond motion generation, practical pipelines include rigging, retargeting, and rendering, making ultra-fast long-term 3D motion generation without compromising quality valuable for real-time interactive applications. Despite promising progress, high-quality long-horizon motion generation under tight inference budgets remains challenging, owing to 2 coupled bottlenecks: (1) generative strategies that require many sampling steps for high-quality 3D motion generation and (2) model architectures with substantial per-step computational and memory costs for long sequences. We present FlashMotion, a co-designed framework for high-quality long-horizon 3D motion generation under a strict inference budget, covering text-to-motion, music-to-dance, and speech-to-gesture tasks. To enhance trajectory-level efficiency, we introduce a 3D motion-space MeanFlow formulation with kinematic and dynamic consistency constraints for few-step generation, together with an Adaptive Soft masking mechanism for temporally coherent motion editing. To improve per-step efficiency, we design a lightweight non-autoregressive BiMamba architecture with condition-guided state selection and AdaIN-based condition fusion mechanism, enhancing cross-modal alignment and generalization to longer motion sequences. Extensive experiments demonstrate that FlashMotion achieves overall state-of-the-art performance in both quality and efficiency across text-to-motion, speech-to-gesture, and music-to-dance generation. The project page is available at https://sun-happy-ykx.github.io/FlashMotion-Project/

cs.CV↗

Möbius: A Unified Motion-Language Model Bridging Understanding and Generation

We aim to build a truly unified motion model that can understand, generate, and edit human motion within a single framework. Existing motion UMMs remain only partially unified: motion understanding is handled by a language model, while generation is often delegated to a separate diffusion module operating over learned motion latents. We introduce Möbius, which brings the entire diffusion process into the same multimodal representation space used for motion understanding. Clean motion, noisy motion, and text jointly participate in shared attention, while role-specific parameters process understanding and generation streams. Importantly, Möbius operates directly on continuous frame-level raw motion, requiring neither discrete tokenization nor a pretrained motion VAE. This simple formulation naturally supports motion captioning, text-to-motion generation, temporal completion, and instruction-guided editing within a single model. On HumanML3D, Möbius achieves state-of-the-art performance among unified motion models. Scaling to 1.7B parameters and 1,700 hours of motion data further demonstrates that our framework is both effective and scalable.

cs.CV↗

WATCH: World-aware Allied Trajectory and pose reConstruction for Camera and Human

Reconstructing global human motion from monocular video is fundamental to VR, graphics, and robotics, yet remains ill-posed due to depth ambiguity, motion ambiguity, and the entanglement of camera and human movements. Human-motion-centric methods achieve strong physical plausibility but leave two signals unused: camera orientation is processed through a fixed coordinate transformation with no independent supervision of its components, and camera velocity is discarded entirely despite being directly observable from SLAM. Camera-trajectory-centric methods use camera translation directly, but hard-decoding SLAM trajectories into human positions propagates depth errors and fails entirely under static cameras. We present WATCH (World-aware Allied Trajectory and pose reConstruction for Camera and Human). The key observation is that once camera orientation is made explicit, camera velocity becomes a natural additional input rather than an ambiguous one. We therefore decompose camera rotation into a network-estimated roll-pitch component and an analytically recoverable yaw, supervising each independently. This decomposition exposes a clean geometric interface through which camera velocity is incorporated as a learned spatial prior in the backbone, without the physically implausible artifacts that arise from hard-decoding. WATCH outperforms prior human-motion-centric methods on both static-camera (RICH) and dynamic-camera (EMDB) benchmarks in global trajectory accuracy, temporal smoothness, and physical plausibility, and remains robust when ground-truth camera is replaced with DPVO estimates.

cs.CV↗

SubtleTalk: Generating Controllable Weakly-correlated Facial Dynamics for 3D Talking Heads via Residual Flow Matching

Audio-driven 3D facial animation aims to synthesize realistic and temporally coherent motions from speech. Despite notable progress in lip synchronization, weakly correlated dynamics, including eyebrow movements, eye blinks, and head motion, which are essential to photorealistic facial animation, remain difficult to model faithfully and often appear static or unnaturally repetitive. We attribute this limitation to three factors: (a) insufficient conditioning for weakly correlated dynamics; (b) the limited ability of deterministic regression to capture diverse motion patterns; (c) data bottlenecks from unreliable upper-face pseudo-labels and limited dataset diversity. To address these issues, we propose SubtleTalk, a framework for generating natural and controllable weakly correlated facial dynamics via multi-condition modeling and residual flow matching. First, to compensate for the limited guidance of speech alone, we introduce interpretable controls, including prosody, regional intensity, and Valence-Arousal signals, to explicitly capture the timing, magnitude, and affective variation of weakly correlated dynamics. Second, to overcome the limited expressiveness of deterministic regression, we build residual flow matching based on a stable speech-driven motion prior, allowing the model to capture stochastic deviations beyond deterministic prediction. Third, to alleviate the data bottleneck, we construct SubtleTalk-Face, a large-scale 3D facial animation dataset comprising about 3,900 identities and 74 hours of data, built via a simple and scalable pseudo-labeling pipeline and featuring improved upper-face tracking and frame-level VA annotations. Extensive experiments demonstrate that our method significantly improves the realism and diversity of weakly correlated facial dynamics while preserving accurate lip synchronization.

cs.GR↗

Real2Sim in HOI: Toward Physically Plausible HOI Reconstruction from Monocular Videos

Recovering 4D human-object interaction (HOI) from monocular video is a key step toward scalable 3D content creation, embodied AI, and simulation-based learning. Recent methods can reconstruct temporally coherent human and object trajectories, but these trajectories often remain visual artifacts while failing to preserve stable contact, functional manipulation, or physical plausibility when used as reference motions for humanoid-object simulation. This reveals a fundamental interaction gap: HOI reconstruction should not stop at tracking a human and an object, but should recover the relation that makes their motion a coherent interaction. We introduce $\textbf{HA-HOI}$, a framework for reconstructing physically plausible 4D HOI animation from in-the-wild monocular videos. Instead of treating the human and object as independent entities in an ambiguous monocular 3D space, we propose a $\textit{human-first, object-follow}$ formulation. The human motion is recovered as the interaction anchor, and the object is reconstructed, aligned, and refined relative to the human action. The resulting kinematic trajectory is then projected into a physics-based humanoid-object simulation, where it acts as a teacher trajectory for stable physical rollout. Across benchmark and in-the-wild videos, $\textbf{HA-HOI}$ improves human-object alignment, contact consistency, temporal stability, and simulation readiness over prior monocular HOI reconstruction methods. By moving beyond visually plausible trajectory recovery toward physically grounded interaction animation, our work takes a step toward turning general monocular HOI videos into scalable demonstrations for humanoid-object behavior. Project page: https://knoxzhao.github.io/real2sim_in_HOI/

cs.CV↗

Animator-Centric Skeleton Generation on Objects with Fine-Grained Details

Skeleton generation is essential for animating 3D assets, but current deep learning methods remain limited: they cannot handle the growing structural complexity of modern models and offer minimal controllability, creating a major bottleneck for real-world animation workflows. To address this, we propose an animator-centric SG framework that achieves high-quality skeleton prediction on complex inputs while providing intuitive control handles. Our contributions are threefold. First, we curate a large-scale dataset of 82,633 rigged meshes with diverse and complicated structures. Second, we introduce a novel semantic-aware tokenization scheme for auto-regressive modeling. This scheme effectively complements purely geometric prior methods by subdividing bones into semantically meaningful groups, thereby enhancing robustness to structural complexity and enabling a key control mechanism. Third, we design a learnable density interval module that allows animators to exert soft, direct control over bone density. Extensive experiments demonstrate that our framework not only generates high-quality skeletons for challenging inputs but also successfully fulfills two critical requirements from professional animators.

cs.GR↗

RoMo: A Robust Solver for Full-body Unlabeled Optical Motion Capture

Optical motion capture (MoCap) is the "gold standard" for accurately capturing full-body motions. To make use of raw MoCap point data, the system labels the points with corresponding body part locations and solves the full-body motions. However, MoCap data often contains mislabeling, occlusion and positional errors, requiring extensive manual correction. To alleviate this burden, we introduce RoMo, a learning-based framework for robustly labeling and solving raw optical motion capture data. In the labeling stage, RoMo employs a divide-and-conquer strategy to break down the complex full-body labeling challenge into manageable subtasks: alignment, full-body segmentation and part-specific labeling. To utilize the temporal continuity of markers, RoMo generates marker tracklets using a K-partite graph-based clustering algorithm, where markers serve as nodes, and edges are formed based on positional and feature similarities. For motion solving, to prevent error accumulation along the kinematic chain, we introduce a hybrid inverse kinematic solver that utilizes joint positions as intermediate representations and adjusts the template skeleton to match estimated joint positions. We demonstrate that RoMo achieves high labeling and solving accuracy across multiple metrics and various datasets. Extensive comparisons show that our method outperforms state-of-the-art research methods. On a real dataset, RoMo improves the F1 score of hand labeling from 0.94 to 0.98, and reduces joint position error of body motion solving by 25%. Furthermore, RoMo can be applied in scenarios where commercial systems are inadequate. The code and data for RoMo are available at https://github.com/non-void/RoMo.

cs.CV↗

EmoFace: Audio-driven Emotional 3D Face Animation

Audio-driven emotional 3D face animation aims to generate emotionally expressive talking heads with synchronized lip movements. However, previous research has often overlooked the influence of diverse emotions on facial expressions or proved unsuitable for driving MetaHuman models. In response to this deficiency, we introduce EmoFace, a novel audio-driven methodology for creating facial animations with vivid emotional dynamics. Our approach can generate facial expressions with multiple emotions, and has the ability to generate random yet natural blinks and eye movements, while maintaining accurate lip synchronization. We propose independent speech encoders and emotion encoders to learn the relationship between audio, emotion and corresponding facial controller rigs, and finally map into the sequence of controller values. Additionally, we introduce two post-processing techniques dedicated to enhancing the authenticity of the animation, particularly in blinks and eye movements. Furthermore, recognizing the scarcity of emotional audio-visual data suitable for MetaHuman model manipulation, we contribute an emotional audio-visual dataset and derive control parameters for each frames. Our proposed methodology can be applied in producing dialogues animations of non-playable characters (NPCs) in video games, and driving avatars in virtual reality environments. Our further quantitative and qualitative experiments, as well as an user study comparing with existing researches show that our approach demonstrates superior results in driving 3D facial models. The code and sample data are available at https://github.com/SJTU-Lucy/EmoFace.

cs.CV↗

SyncDreamer: Generating Multiview-consistent Images from a Single-view Image

In this paper, we present a novel diffusion model called that generates multiview-consistent images from a single-view image. Using pretrained large-scale 2D diffusion models, recent work Zero123 demonstrates the ability to generate plausible novel views from a single-view image of an object. However, maintaining consistency in geometry and colors for the generated images remains a challenge. To address this issue, we propose a synchronized multiview diffusion model that models the joint probability distribution of multiview images, enabling the generation of multiview-consistent images in a single reverse process. SyncDreamer synchronizes the intermediate states of all the generated images at every step of the reverse process through a 3D-aware feature attention mechanism that correlates the corresponding features across different views. Experiments show that SyncDreamer generates images with high consistency across different views, thus making it well-suited for various 3D generation tasks such as novel-view-synthesis, text-to-3D, and image-to-3D.

cs.CV↗

Motion-R3: Fast and Accurate Motion Annotation via Representation-based Representativeness Ranking

In this paper, we follow a data-centric philosophy and propose a novel motion annotation method based on the inherent representativeness of motion data in a given dataset. Specifically, we propose a Representation-based Representativeness Ranking R3 method that ranks all motion data in a given dataset according to their representativeness in a learned motion representation space. We further propose a novel dual-level motion constrastive learning method to learn the motion representation space in a more informative way. Thanks to its high efficiency, our method is particularly responsive to frequent requirements change and enables agile development of motion annotation models. Experimental results on the HDM05 dataset against state-of-the-art methods demonstrate the superiority of our method.

cs.CV↗

Diverse Motion In-betweening with Dual Posture Stitching

In-betweening is a technique for generating transitions given initial and target character states. The majority of existing works require multiple (often $>$10) frames as input, which are not always accessible. Our work deals with a focused yet challenging problem: to generate the transition when given exactly two frames (only the first and last). To cope with this challenging scenario, we implement our bi-directional scheme which generates forward and backward transitions from the start and end frames with two adversarial autoregressive networks, and stitches them in the middle of the transition where there is no strict ground truth. The autoregressive networks based on conditional variational autoencoders (CVAE) are optimized by searching for a pair of optimal latent codes that minimize a novel stitching loss between their outputs. Results show that our method achieves higher motion quality and more diverse results than existing methods on both the LaFAN1 and Human3.6m datasets.

cs.CV↗