MorphGS: Morphology-Adaptive Articulated 3D Motion Transfer from Videos
This addresses the problem of motion retargeting for 3D animation, offering a more flexible approach compared to existing methods that rely on intermediate reconstruction, though it appears incremental as it builds on prior analysis-by-synthesis techniques.
The paper tackles the challenge of transferring articulated motion from monocular videos to rigged 3D characters by proposing MorphGS, a framework that directly optimizes target morphology and pose using image-space supervision, resulting in consistent improvements over baselines in experiments on synthetic benchmarks and in-the-wild videos.
Transferring articulated motion from monocular videos to rigged 3D characters is challenging due to pose ambiguity in 2D observations and morphological differences between source and target. Existing approaches often follow a reconstruct-then-retarget paradigm, tying transfer quality to intermediate 3D reconstruction and limiting applicability to categories with parametric templates. We propose MorphGS, a framework that formulates motion retargeting as a target-driven analysis-by-synthesis problem, directly optimizing target morphology and pose through image-space supervision. A rig-coupled morphology parameterization factorizes character identity from time-varying joint rotations, while dense 2D-3D correspondences and synthesized views provide complementary structural and multi-view guidance. Experiments on synthetic benchmarks and in-the-wild videos show consistent improvements over baselines.