The paper introduces FreqMo, a motion representation that splits human motion into wavelet frequency bands to handle varying temporal scales. By using Unified Frequency Residual Quantization, it compresses tokens threefold and supports stable single-stage generation. Experiments demonstrate state‑of‑the‑art fidelity, especially in preserving high‑frequency motion details, and the method is compatible with continuous diffusion models.
arXiv:2608.23279v1 Announce Type: new Abstract: Text-driven human motion synthesis has made substantial development with two core modules of motion representation and generative architecture. For rep...
arXiv:2608.29891v1 Announce Type: new Abstract: Unsupervised skeleton-based temporal action segmentation is a crucial task for understanding human behavior in long untrimmed sequences. Recent approac...
arXiv:2605. 16366v2 Announce Type: replace-cross Abstract: Video MLLMs face a persistent tension between spatial fidelity and temporal coverage: preserving fine-grained visual details requires many spatial tokens, while capturing short-lived events requires dense temporal sampling.
The paper introduces Frequency-Factorized Multi-Rate Tokenization (FFMT) for music-to-dance generation, separating root-relative motion into a low-frequency choreographic scaffold and a high-frequency residual. FFMT assigns a lower token rate and smaller codebook to the low-frequency component, while using a higher token rate and larger codebook for the high-frequency part, enabling a plan-then-articulate autoregressive factorization. Experiments on AIST++ and FineDance demonstrate strong distributional fidelity, realistic diversity, and music–motion synchronization, with ablations confirming the benefits of frequency factorization, asymmetric temporal allocation, and low-to-high conditioning.
arXiv:2608.24334v1 Announce Type: new Abstract: Discrete motion representations have substantially advanced autoregressive text-to-motion generation. However, most motion tokenizers are optimized for...
MotionSpec introduces a motion supervision framework for text-to-video generation that focuses on Spectral Trajectory Consistency (STC). STC builds dense anchor-relative motion trajectories, transforms them into spectral volumes, and aligns their amplitude and phase with target trajectories to constrain motion strength and temporal organization. The framework also adds Local Flow Consistency (LFC) to stabilize local motion transitions, resulting in improved motion consistency, temporal coherence, and plausibility while maintaining visual fidelity.
The paper introduces a latent dataset distillation framework for human motion prediction, addressing the limitations of traditional gradient matching by incorporating a learned motion prior. Motions are compressed using a residual‑quantized variational autoencoder, and distillation updates only a latent bank while keeping the decoder frozen, ensuring synthetic motions remain plausible. Experiments on Human3.6M, CMU, and 3DPW datasets demonstrate that this method outperforms direct gradient matching in most settings and yields more realistic synthetic motions.
4D Gaussian Splatting (4DGS) excels in dynamic 3D reconstruction and real-time novel view synthesis via efficient 4D Gaussian representations and parallelizable rendering. However, existing 4DGS approaches rely on a single polynomial to model motion, which limits performance in complex dynamic scenes where high-frequency motion components are prevalent, and fails to ensure long-term stability due to cumulative trajectory drift.
The paper introduces MOCO, a diffusion-based framework that generates 3D avatar motions from concurrent multimodal inputs such as speech audio, text descriptions, and trajectory data. MOCO decouples motion generation by independently producing modality-specific motions at each denoising step and then assembling them according to spatial rules, iteratively refining the combined motion. This approach yields coherent, lifelike, and synchronized movements, outperforming existing baselines on a multimodal benchmark.
Uni-HOI is a unified framework that learns the joint distribution among text, human motion, and object motion for 4D human‑object interaction (HOI). It uses large language models and two motion‑specific VQ‑VAEs to convert heterogeneous motion data into token sequences, enabling seamless integration of all three modalities. A two‑stage training strategy first captures correlations on a large‑scale HOI dataset and then fine‑tunes for specific tasks, achieving strong performance on text‑driven HOI generation, object‑motion‑driven human motion generation, and human‑motion‑driven object motion prediction.
arXiv:2608. 01958v2 Announce Type: replace-cross Abstract: 4D Gaussian Splatting (4DGS) excels in dynamic 3D reconstruction and real-time novel view synthesis via efficient 4D Gaussian representations and parallelizable rendering.
arXiv:2605. 29488v2 Announce Type: replace-cross Abstract: Conditional human motion generation remains a fundamental challenge in computer vision and robotics.