Jul 2026· ACM Transactions on Graphics· 0 citations· 24 references
TL;DR
This work proposes Directable Motion Paraphrasing (DMP), a novel motion retargeting framework based on the concept of motion paraphrasing, analogous to text paraphrasing, where the core semantics of a motion are preserved while allowing expressive, user-directed variations.
Abstract
Creating diverse and realistic human motions is a fundamental cornerstone of computer animation, with numerous applications in games, movies, and AR/VR. While motion capture is a valuable tool for capturing motions across varied body sizes, obtaining unique motion data for a variety of characters is often prohibitively expensive. Motion retargeting addresses this limitation by adapting existing motions to different character morphologies, however, existing approaches often involve trade-offs between motion realism, user control, and adaptability to artistic needs. In this work, we propose Directable Motion Paraphrasing (DMP), a novel motion retargeting framework based on the concept of motion paraphrasing, analogous to text paraphrasing, where the core semantics of a motion are preserved while allowing expressive, user-directed variations. Our framework constructs a large-scale motion paraphrasing dataset, which captures the diversity of human motion across different body shapes, and trains a diffusion-based generative model that learns both invariances and variations in motion. To enable user control during inference, we introduce a flexible mechanism for specifying spatio-temporal constraints, such as joint positions, rotations, and object interactions, which can be incorporated into the generative process through masked inpainting and loss guidance. We demonstrate the effectiveness of our framework through various examples, showing its ability to produce realistic, diverse, and controllable retargeted motions that meet the artistic demands of animation pipelines. Extensive experiments demonstrate the system’s flexibility, motion plausibility, and directability, highlighting its potential as a tool for intuitive and high-quality motion retargeting.
Realistic animal motion for virtual production is typically obtained either through motion capture of highly trained performers who accurately mimic animal behavior, or by retargeting ordinary human motion using complex control setups. Both approaches are challenging and often fail to fully reproduce the nuances of natural animal motion, motivating data-driven alternatives. We present an automatic human-to-quadruped puppeteering framework that produces plausible and controllable quadruped motions from ordinary human motion data. Our approach employs a two-stage generative diffusion model trained purely on quadruped motion data. By introducing a structured conditioning and inpainting strategy, our method supports a wide range of actions, including walking, running, jumping, sitting, and lying. Furthermore, we enable fine-grained intuitive control of the quadruped motion such as head movement control and individual limb puppeteering. Experimental results demonstrate improved motion realism and controllability compared to existing retargeting approaches, highlighting the effectiveness of our framework as a tool for animation and virtual production applications.
Fatemeh Zargarbashi, Zehong Qiu, Dhruv Agrawal et al.· 0 citations
This work explores the motion transfer from one video to another, which is crucial in animation for diverse characters. Previously, video motion transfer has been largely explored between human and human-like characters, enabling a lot of applications in digital creation. However, these approaches encounter a main limitation. Specifically, related technical pipelines heavily rely on a predefined human skeleton structure and accordingly require skeleton-conditional model training. On the one hand, these methods are difficult to generalize to diverse characters, such as animals from different species, while preserving their unique motion styles. On the other hand, labeled data in diverse skeletons is limited, which additionally restricts the large-scale training for the task. In this paper, we jump out of the skeleton-based motion transfer framework and propose a training-free motion transfer framework, named Motion4Motion. Motion4Motion models the motion flow of the character in a video instead of skeletons, which makes motion transfer across species easier. Extensive experimental results and novel applications show our methods outperform baselines impressively.
Ling-Hao Chen, Zixin Yin, Duomin Wang et al.· International Conference on...· 0 citations
The recent success of generative modeling has led to entirely new capabilities to author 3D motion. By manipulating only a few sparse handles and poses, it is now possible to generate entire motion sequences, at scale and with many variations. To bridge the gap between model research and real-world integration, we developed a Generative Motion Rig as a Blender plugin, built atop a general motion model. Our rig supports a new “generative keyframing” workflow, where artists author movements by manipulating sparse poses, handles, window lengths, and noise sampling. We show in our accompanying video how artists use these controls to rapidly make a short animation. We also show how the same rig can support generative motion editing, allowing one to edit and extend mocap clips. Finally, we share insights and future challenges to help close the gap between generative and traditional animation workflows.
Jakob Buhmann, Dhruv Agrawal, D. Borer et al.· Proceedings of the Special I...· 0 citations
In combat sports simulation, expert users have extensive experience in extracting visual anticipatory information from their opponent’s movements, which should be preserved in virtual opponent animations. However, animations based on motion capture data rely on complex pipelines that may disrupt the visual information initially present in the real actor’s motion. Recent physics-based simulation (such as Adversarial Motion Prior, AMP) have recently been introduced to adapt character motion to interactive constraints while satisfying physical laws, but they may introduce additional distortions. In this paper, we examine how animating a virtual opponent using motion capture data (as a reference), retargeting techniques, and AMP influences the visual anticipatory performance of expert boxers. Ten boxers participated in a perceptual study using video clips with varying occlusion times, in which they were asked to predict the type of punch among four possible choices. The results show a clear decline in anticipation perceptual accuracy (PA) as occlusion occurs earlier. Hooks were more difficult to anticipate than straight punches (PA for rear hook = 0.455, lead hook = 0.435, rear straight = 0.660, lead straight = 0.744). Animations based on motion capture (mean accuracy = 0.758) led to significant better anticipation perceptual accuracy, for straight punches, compared to retargeting (mean accuracy = 0.700) and AMP (mean accuracy = 0.647). Overall, these results highlight the need to redesign traditional animation pipelines to better preserve the visual information picked up by expert users in virtual training simulators.
Annabelle Limballe, Ahmed Abdourahman Mahamoud, Guillaume Claude et al.· Proceedings of the 2026 ACM...· 0 citations
Motion warping is a core technique in character animation that enables the adaptation of existing motion data to novel spatio-temporal constraints. Conventional motion warping methods often rely on heuristic modifications that can violate physical consistency or introduce visual artifacts. More recent learning-based editing approaches improve realism, but many of them encode motion into tightly entangled latent space, which makes them struggle to balance editing flexibility and content preservation. To address this, we propose a novel deep motion warping framework that explicitly disentangles the motion structure from global and stylistic attributes for intuitive motion editing. Our key insight is to leverage learned phase features as a continuous and robust representation of the underlying structure, and explicitly disentangle motion into root velocity, phase, and learned latent variables using a phase-conditioned diffusion autoencoder. This design supports a wide range of editing operations, including root motion warping, motion exaggeration, time warping, and style transfer by directly manipulating decoupled components, without requiring paired training data. Extensive experiments demonstrate that our approach enables high-level, flexible motion editing while strictly preserving the structural consistency and physical plausibility of the source motion
Bowen Zheng, Linjun Wu, Xinwei Jiang et al.· International Conference on...· 0 citations
Motion editing is central to VFX and game development, where it is used extensively to modify and augment existing movements to conform to new environments or changes in artistic direction. While traditional motion editing can do small modifications, it cannot accommodate larger structural edits, resulting in visual warping artifacts that require authoring new motion. Conversely, recent advances in large-scale generative modeling have unlocked newfound capabilities for authoring entire movements by directly manipulating sparse spatial constraints. While impressive at creating new movements, these methods lack the capability to preserve and edit existing motion interactively. In this work, we introduce scheduled inpainting, a method that enables interactive generative motion editing, a novel paradigm unifying motion synthesis and editing by leveraging generative models. Scheduled inpainting is a simple yet powerful inference-based technique that enables fine-grained spatiotemporal control over the balance between preserving the original motion and generating new content. By building atop generative models that support direct manipulation, our system allows artists to interactively refine existing animations while ensuring results remain natural and consistent with the learned motion distribution. Scheduled inpainting is versatile and supports many editing applications, such as extending, stitching, and compositing different clips. Finally, we extensively validate our approach by comparing with four baselines, conducting ablations of our design, and reporting user feedback.
Dhruv Agrawal, D. Borer, Luca Vögeli et al.· 0 citations