Research map / 3D vision / Face and shape
CARTO: Category and Joint Agnostic Reconstruction of ARTiculated Objects
Nick Heppert, Muhammad Zubair Irshad, Sergey Zakharov, Katherine Liu, Rares Andrei Ambrus, Jeannette Bohg, Abhinav Valada, Thomas Kollar
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use a single-shot stereo-image pipeline instead of the two-stage A-SDF pipeline | CARTO · Full scene reconstruction, synthetic test set | mAP IOU50: 10.4 -> 31.5 | Improved |
| use a single-shot stereo-image pipeline instead of the two-stage A-SDF pipeline | CARTO · Full scene reconstruction, synthetic test set | mAP IOU25: 33.9 -> 64.0 | Improved |
| use a single-shot stereo-image pipeline instead of the two-stage A-SDF pipeline | CARTO · Full scene reconstruction, synthetic test set | 20°30cm pose accuracy: 70.8 -> 76.6 | Improved |
| use a single-shot stereo-image pipeline instead of the two-stage A-SDF pipeline | CARTO · Full scene reconstruction, two-object inference | Total time: 64.262 -> 1.092 | Improved |
| add joint-space similarity regularization, including pre-training and post-epoch optimization (ablation) | CARTO decoder · Canonical reconstruction task, seven-category mean | Chamfer distance (CD): 2.246 -> 1.192 | Improved |
| add joint-space similarity regularization, including pre-training and post-epoch optimization (ablation) | CARTO decoder · Canonical reconstruction task, revolute categories mean | Joint state error: 35.892° -> 11.512° | Improved |
| add joint-space similarity regularization, including pre-training and post-epoch optimization (ablation) | CARTO decoder · Canonical reconstruction task, seven-category mean | Joint type accuracy: 0.646 -> 0.908 | Improved |
| train CARTO jointly across categories rather than only on one category | CARTO decoder · Canonical reconstruction task, seven-category mean | Chamfer distance (CD): 1.190 -> 1.192 | No clear effect |
| train CARTO jointly across categories rather than only on one category | CARTO decoder · Canonical reconstruction task, revolute categories mean | Joint state error: 12.474° -> 11.512° | Mixed |
| replace stereo RGB with RGB-D input (ablation) | CARTO-D · Full scene reconstruction, synthetic test set | mAP IOU50: 31.5 -> 38.2 | Improved |
| replace stereo RGB with RGB-D input (ablation) | CARTO-D · Real-world test dataset | meaningful detections | Hurt |
| apply A-SDF test-time adaptation by jointly optimizing the shape code, joint state, and network weights | A-SDF · Canonical reconstruction task | Chamfer distance: 13.339 | Mixed |
Later papers that compared with this method
2 later accepted papers report CARTO as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- SIAM: Towards Generalizable Articulated Object Modeling via Single Robot-Object Interaction (AAAI 2026 · better on 1 of 1 shared result)
- Category-Level Articulated Object Pose Estimation via Pose–Shape Hypothesis Generation and Verification (ECCV 2026 · better on 1 of 1 shared result)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- A-SDF — A-SDF: Learning Disentangled Signed Distance Functions for Articulated Shape Representation (ICCV 2021) · better on 5 of 7 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Real2Code: Reconstruct Articulated Objects via Code Generation (ICLR 2025 · similarity 0.88)
- REACTO: Reconstructing Articulated Objects from a Single Video (CVPR 2024 · similarity 0.87)
- DIPO: Dual-State Images Controlled Articulated Object Generation Powered by Diverse Data (NeurIPS 2025 · similarity 0.87)
- Learning Implicit Representation for Reconstructing Articulated Objects (ICLR 2024 · similarity 0.87)
- S3O: A Dual-Phase Approach for Reconstructing Dynamic Shape and Skeleton of Articulated Objects from Single Monocular Video (ICML 2024 · similarity 0.86)
- SINGAPO: Single Image Controlled Generation of Articulated Parts in Objects (ICLR 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
