Human Motion Generation
Overview
Human motion generation studies how to synthesize or edit body movement for animation, avatars, interaction, and humanoid robots. Common approaches use diffusion or flow-matching models, autoregressive and masked motion tokens, and structured representations of body parts or joint trajectories; retrieval, planning, and physics-based refinement can add control. Text-to-motion is often tested on HumanML3D and KIT-ML, while dance and speech-driven gestures use benchmarks such as AIST++ and BEAT. A central challenge is balancing semantic accuracy, natural and diverse motion, physical plausibility, and precise control—especially for long or real-time sequences, where evaluation metrics may not reflect human judgments or successful execution.
Latest papers
309 papers20 papers in the last four weeks, down 47% on the four weeks before. 0.2% of all new papers.
InterCMDM: Block-Causal Diffusion for Autoregressive Human Interaction Generation
LUNA: Learning Universal 3D Human Animation Beyond Skinning
PoseShield: Neural Collision Fields for Human Self-Collision Resolution
GPC: Large-Scale Generative Pretraining for Transferable Motor Control
Generative Learning as a Tool to Improve Perception of Emotional Body Motion Expressions
EMOSH: Expressive Motion and Shape Disentanglement for Human Animation
Beyond MoCap: Scaling Motion Tokenizers with Synthetic Human Motion for Generative Modeling
In-Context Model Predictive Generation: Open-Vocabulary Motion Synthesis from Language Models to Physics
DanceDuo: Bridging Human Movement and AI Choreography
Follow Your Track: Precise Skeleton Animation Controlled by 3D Trajectories
ACT, a trajectory-conditioned framework for topology-general skeletal animation. ACT uses skeletons as a compact structured and compute-efficient representation and 3D point trajectories from monocular video as explicit motion guidance which provide detailed motion patterns without appearance entanglement. At the core of ACT is a Routed Trajectory Injector, which achieves accurate and robust trajectory-to-joint transfer through three complementary designs: prior-guided hard routing establishes precise skeleton-to-mesh correspondences, global routing enables holistic joint-track interaction for full-body motion awareness, and local windowed cross-attention enforces fine-grained temporal alignment, improving micro-timing and reducing motion misalignment across varying motion rates. Extensive experiments demonstrate that ACT significantly outperforms existing methods in fidelity and temporal consistency.Social Structure Matters in 3D Human-Human Interaction Generation
IMAGIN-4D: Image-Guided Controllable Interaction Generation
TEXEDO : Test Time Scaling for Controller-aware Language-conditioned Humanoid Motion Generation
Text Dictates, Music Decorates: Energy-based Attention for Editable Dance Motion Generation
Context-Aware Autoregressive Diffusion for Gloss-Wise Sign Language Production
Odoriko: A Shape-Aware Multimodal Diffusion Framework for Human Motion
PhysDrift: Bridging the Embodiment Gap in Humanoid Co-Speech Motion Generation
XmoPipe: A Pipeline for Large-Scale In-the-Wild Human Motion Dataset Construction
Generating Natural and Expressive Robot Gestures through Iterative Reinforcement Learning with Human Feedback using LLMs
MOCHI: Motion Enhancement of Collaborative Human-object Interactions
WaveSync: Constrained Wavefront Optimization for Synchronized Co-Speech Gestures in Humanoid Robots
SiGnature: Explicit Motion Diffusion for Stylized Semantic Gesture
active joints'' conveying a semantic action and injects them into the generation, while relying on the diffusion backbone to synthesize the remaining body dynamics, including posture and flow, in accordance with the pre-learned style of the target speaker. This allows for the plug-and-play integration of arbitrary motions, including complex semantic gestures, without retraining or introducing the Frankenstein'' artifacts typical of cut-and-paste methods. Extensive experiments and perceptual studies demonstrate that SiGnature offers superior semantic motion control while maintaining smooth and natural co-speech gesture generation and preserving the distinct characteristics of the speaker, thereby outperforming state-of-the-art baselines.