← ArchivePaper2024
Geometry-Aware Retargeting for Two-Skinned Characters Interaction
Inseo Jang, Soojin Choi, Seokhyeon Hong, Chaelin Kim, Junyong Noh
Retargets interaction motions between two characters while preserving skinned mesh geometry and contact semantics.
Abstract
Interactive motion between multiple characters is widely utilized in games and movies. However, the method for generating interactive motions considering the character's diverse mesh shape has yet to be studied. We propose a Spatio Cooperative Transformer (SCT) to retarget the interacting motions of two characters having arbitrary mesh connectivity. SCT predicts the residual of root position and joint rotations considering the shape difference between the source and target of interacting characters. In addition, we introduce an anchor loss function for SCT to maintain the geometric distance between the interacting characters when they are retargeted. We also propose a motion augmentation method with deformation-based adaptation to prepare a source-target paired dataset with an identical mesh connectivity for training. In experiments, our method achieved higher accuracy for semantic preservation and produced less artifacts of inter-penetration between the interacting characters for unseen characters and motions than the baselines. Moreover, we conducted a user evaluation using characters with various shapes, spanning low-to-high interaction levels to prove better semantic preservation of our method compared to previous studies.
How to read this
- Category
- Method: interaction-aware motion retargeting
- Contributions
- A Spatio Cooperative Transformer (SCT) that retargets the interacting motions of two characters with arbitrary mesh connectivity by predicting residual root position and joint rotations from shape differences.
- An anchor loss that maintains the geometric distance between interacting characters after retargeting, reducing inter-penetration.
- A motion augmentation with deformation-based adaptation to build a source-target paired dataset with identical mesh connectivity for training.
- Context
- Extends contact-aware skinned-motion retargeting (Villegas et al., 'Contact-Aware Retargeting of Skinned Motion') to the two-character interaction setting where both skinned shapes must be respected.Builds on: Contact-Aware Retargeting of Skinned Motion
- Correctness
- Reported to give higher semantic preservation and fewer inter-penetration artifacts than baselines on unseen characters and motions, plus a user study; the validation rests on its augmented paired dataset, so generalization beyond the trained interaction types is a reasonable caution.
- Clarity
- Accessible problem statement; a first pass conveys the SCT and anchor-loss idea, a second pass for the residual prediction and augmentation pipeline.
- How to read it
- First pass for the two-character contact-preservation goal and the anchor loss; second pass on the SCT architecture and dataset augmentation if you work on retargeting interacting characters.
Builds on
Built upon by
Nothing yet.
Related work
- Contact-Aware Retargeting of Skinned Motion 2021 / CVPR
- Aura Mesh: Motion Retargeting to Preserve the Spatial Relationships between Skinned Characters 2018 / CGF
- Normalized Euclidean Distance Matrices for Human Motion Retargeting 2017 / MIG
- Dog Code: Human to Quadruped Embodiment Using Shared Codebooks 2024 / MIG
Keywords
This page summarises the entry and links to its original source. The archive never hosts or redistributes the publication itself.Show it in the full archive list →