REVIEW 6 cited by
Pay Attention and Move Better: Harnessing Attention for Interactive Motion Generation and Training-free Editing
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
This research delves into the problem of interactive editing of human motion generation. Previous motion diffusion models lack explicit modeling of the word-level text-motion correspondence and good explainability, hence restricting their fine-grained editing ability. To address this issue, we propose an attention-based motion diffusion model, namely MotionCLR, with CLeaR modeling of attention mechanisms. Technically, MotionCLR models the in-modality and cross-modality interactions with self-attention and cross-attention, respectively. More specifically, the self-attention mechanism aims to measure the sequential similarity between frames and impacts the order of motion features. By contrast, the cross-attention mechanism works to find the fine-grained word-sequence correspondence and activate the corresponding timesteps in the motion sequence. Based on these key properties, we develop a versatile set of simple yet effective motion editing methods via manipulating attention maps, such as motion (de-)emphasizing, in-place motion replacement, and example-based motion generation, etc. For further verification of the explainability of the attention mechanism, we additionally explore the potential of action-counting and grounded motion generation ability via attention maps. Our experimental results show that our method enjoys good generation and editing ability with good explainability.
Forward citations
Cited by 6 Pith papers
-
Absolute Coordinates Make Motion Generation Easy
Using absolute 3D joint coordinates with a plain Transformer and velocity-prediction diffusion outperforms the standard local-relative motion representation, improving fidelity and enabling direct control.
-
Interactive Generative Motion Editing via Scheduled Inpainting
Scheduled inpainting blends a base motion clip into a diffusion model's denoising process via a user-controlled schedule and spatiotemporal mask, enabling interactive editing of existing animations without retraining.
-
Motion4Motion: Motion Transfer Across Subjects at Inference
Training-free motion transfer across species works by extracting source motion flows, matching semantic points, and injecting them into DiT self-attention via TransPE positional padding.
-
Go to Zero: Towards Zero-shot Motion Generation with Million-scale Data
A 7B text-to-motion model trained on the new 2M-clip MotionMillion dataset is reported to generalize zero-shot to complex, out-of-domain prompts.
-
CoDA: Coordinated Diffusion Noise Optimization for Whole-Body Manipulation of Articulated Objects
CoDA generates coordinated whole-body articulated-object manipulation by optimizing the noise of three decoupled diffusion models, guided by BPS-based end-effector and object trajectories.
-
From Motion to Behavior: Hierarchical Modeling of Humanoid Generative Behavior Control
A new 124K-clip dataset with hierarchical text annotations, plus a pipeline that couples an LLM planner, a text-to-pose VAE, diffusion in-betweening, and physics control to generate long-horizon human behaviors.
Discussion (0). Continue with ORCID to comment.