# SCAPO: Self-Supervised Category-Level Articulated Pose Estimation from a Single 3D Observation

arXiv: https://arxiv.org/abs/2606.01940 · HTML: https://mlsanity.com/paper/2606.01940/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 26 of its 31 references that are in the map.
- First posted on arXiv 1 Jun 2026.

## Builds on (most cited first, 26 of 26)

- [HOI4D: A 4D Egocentric Dataset for Category-Level Human-Object Interaction](https://mlsanity.com/paper/2203.01577/) (2022)
- [Vector Neurons: A General Framework for SO(3)-Equivariant Networks](https://mlsanity.com/paper/2104.12229/) (2021)
- [BANMo: Building Animatable 3D Neural Models from Many Casual Videos](https://mlsanity.com/paper/2112.12761/) (2021)
- [Neural Articulated Radiance Field](https://mlsanity.com/paper/2104.03110/) (2021)
- [Category-Level Articulated Object Pose Estimation](https://mlsanity.com/paper/1912.11913/) (2019)
- [Ditto: Building Digital Twins of Articulated Objects from Interaction](https://mlsanity.com/paper/2202.08227/) (2022)
- [NASA: Neural Articulated Shape Approximation](https://mlsanity.com/paper/1912.03207/) (2019)
- [Shape2Motion: Joint Analysis of Motion Parts and Attributes from 3D Shapes](https://mlsanity.com/paper/1903.03911/) (2019)
- [PARIS: Part-level Reconstruction and Motion Analysis for Articulated Objects](https://mlsanity.com/paper/2308.07391/) (2023)
- [A-SDF: Learning Disentangled Signed Distance Functions for Articulated Shape Representation](https://mlsanity.com/paper/2104.07645/) (2021)
- [LEAP: Learning Articulated Occupancy of People](https://mlsanity.com/paper/2104.06849/) (2021)
- [Neural Parts: Learning Expressive 3D Shape Abstractions with Invertible Neural Networks](https://mlsanity.com/paper/2103.10429/) (2021)
- [CAPTRA: CAtegory-level Pose Tracking for Rigid and Articulated Objects from Point Clouds](https://mlsanity.com/paper/2104.03437/) (2021)
- [Equivariant Point Network for 3D Point Cloud Analysis](https://mlsanity.com/paper/2103.14147/) (2021)
- [Neural Implicit Representation for Building Digital Twins of Unknown Articulated Objects](https://mlsanity.com/paper/2404.01440/) (2024)
- [CARTO: Category and Joint Agnostic Reconstruction of ARTiculated Objects](https://mlsanity.com/paper/2303.15782/) (2023)
- [CaDeX: Learning Canonical Deformation Coordinate Space for Dynamic Surface Representation via Neural Homeomorphism](https://mlsanity.com/paper/2203.16529/) (2022)
- [CLA-NeRF: Category-Level Articulated Neural Radiance Field](https://mlsanity.com/paper/2202.00181/) (2022)
- [Self-supervised Neural Articulated Shape and Appearance Models](https://mlsanity.com/paper/2205.08525/) (2022)
- [REACTO: Reconstructing Articulated Objects from a Single Video](https://mlsanity.com/paper/2404.11151/) (2024)
- [ArticulatedGS: Self-supervised Digital Twin Modeling of Articulated Objects using 3D Gaussian Splatting](https://mlsanity.com/paper/2503.08135/) (2025)
- [Self-Supervised Category-Level Articulated Object Pose Estimation with Part-Level SE(3) Equivariance](https://mlsanity.com/paper/2302.14268/) (2023)
- [Self-Supervised Learning of Part Mobility from Point Cloud Sequence](https://mlsanity.com/paper/2010.11735/) (2020)
- [Shape-Pose Disentanglement using SE(3)-equivariant Vector Neurons](https://mlsanity.com/paper/2204.01159/) (2022)
- [IAAO: Interactive Affordance Learning for Articulated Objects in 3D Environments](https://mlsanity.com/paper/2504.06827/) (2025)
- [OP-Align: Object-level and Part-level Alignment for Self-supervised Category-level Articulated Object Pose Estimation](https://mlsanity.com/paper/2408.16547/) (2024)

## Abstract (arXiv)

Existing methods for category-level object articulation from a single 3D observation often rely on dense supervision, multi-frame inputs, or CAD templates, and still struggle to disentangle geometry from articulation or to recover explicit joint parameters. We propose SCAPO, a self-supervised framework that estimates canonical geometry, rigid part segmentation, and joint pivots, axes, and articulation states from a single RGB-D observation without ground-truth labels or category-specific models. Our SCAPO first uses an SE(3)-equivariant vector-neuron autoencoder to factor out global pose and align diverse instances into a shared canonical space. On this aligned shape, a joint-aware blend-skinning module is then designed to model part motion. We learn this representation through cycle reconstruction between observed and canonical shapes and cross-space alignment with a learnable canonical template that decouples shared category geometry from instance-specific residual shape. Experiments on synthetic and real articulated-object datasets show that our SCAPO recovers consistent part structure and accurate articulation parameters and outperforms all self-supervised baselines.