😸 MimiCAT: Mimic with Correspondence-Aware Cascade-Transformer for Category-Free 3D Pose Transfer

CVPR, 2026
Zenghao Chai1, Chen Tang2, Yongkang Wong1, Xulei Yang3, Mohan Kankanhalli1,
1School of Computing, National University of Singapore,
2MMLab, The Chinese University of Hong Kong, 3Institute for Infocomm Research, A*STAR
Paper arXiv Video (TBD) Data (TBD)

Overview

Overview of MimiCAT for category-free pose transfer.

MimiCAT takes a paired source pose and target character as input. It first employs the correspondence transformer 𝒢 to estimate soft keypoint correspondences, then refines the initialized transformations using the pose transfer transformer 𝒣 to generate the target transformations. Finally, the target character is deformed into the desired pose through linear blend skinning (LBS).

 

MimiCAT for category-free 3D pose transfer.

Given source character with desired poses (left), our model faithfully transfers the given pose to the target characters (right) across completely different categories, proportions and topologies, without requirement of manually labeled correspondence.

Abstract

3D pose transfer aims to transfer the pose-style of a source mesh to a target character while preserving both the target's geometry and the source's pose characteristic. Existing methods are largely restricted to characters with similar structures and fail to generalize to category-free settings (e.g., transferring a humanoid's pose to a quadruped). The key challenge lies in the structural and transformation diversity inherent in distinct character types, which often leads to mismatched regions and poor transfer quality.

To address these issues, we first construct a million-scale pose dataset across hundreds of distinct characters. We further propose To address these issues, we first construct a million-scale pose dataset across hundreds of distinct characters. We further propose MimiCAT, a cascade-transformer model designed for category-free 3D pose transfer. Instead of relying on strict one-to-one correspondence mappings, MimiCAT leverages semantic keypoint labels to learn a novel soft correspondence that enables flexible many-to-many matching across characters. The pose transfer is then formulated as a conditional generation process, in which the source transformations are first projected onto the target through soft correspondence matching and subsequently refined using shape-conditioned representations. Extensive qualitative and quantitative experiments demonstrate that MimiCAT generalizes plausible poses across diverse character morphologies, surpassing prior approaches restricted to narrow-category transfer (e.g., humanoid-to-humanoid)., a cascade-transformer model designed for category-free 3D pose transfer. Instead of relying on strict one-to-one correspondence mappings, MimiCAT leverages semantic keypoint labels to learn a novel soft correspondence that enables flexible many-to-many matching across characters. The pose transfer is then formulated as a conditional generation process, in which the source transformations are first projected onto the target through soft correspondence matching and subsequently refined using shape-conditioned representations.

Extensive qualitative and quantitative experiments demonstrate that MimiCAT generalizes plausible poses across diverse character morphologies, surpassing prior approaches restricted to narrow-category transfer (e.g., humanoid-to-humanoid).

Video

TBD

Pipeline

Overview of the correspondence transformer 𝒢. We (a) first extract shape and keypoint tokens using the shape projector and keypoint encoder, (b) fuse shape conditions with respective keypoint latents through transformer blocks, (c) estimate correspondences via learnable affinity weights followed by the Sinkhorn algorithm, and (d) produce soft-matching correspondences between the given characters.

 

Overview of the correspondence transformer ℋ. We (a) first perform cross-attention to extract deformation-aware cues for shape tokenization and apply correspondence-aware initialization for keypoint tokenization. (b) The shape and keypoint tokens are fed into transformer blocks to derive high-level representations, and decode into refined target transformations. (c) the posed target mesh is generated by deforming the canonical target through Eq. 1.

Dataset: PokeAnimDB

We collect a dataset comprises hundreds of characters spanning a broad spectrum of species and morphologies, including humanoids, quadrupeds, birds, reptiles, fishes, and insects. Each character is paired with artist-designed skeletal animations, resulting in a total of 28k motions and 4.4 million frames.


BibTeX

@article{chai2026mimicat,
  author    = {Chai, Zenghao and Tang, Chen and Wong, Yongkang and Yang, Xulei and Kankanhalli, Mohan},
  title     = {MimiCAT: Mimic with Correspondence-Aware Cascade-Transformer for Category-Free 3D Pose Transfer},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year      = {2026},
}