Research map / 3D vision / Human pose estimation
SPEC: Seeing People in the Wild With an Estimated Camera
Muhammed Kocabas, Chun-Hao P. Huang, Joachim Tesch, Lea Müller, Otmar Hilliges, Michael J. Black
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 |
|---|---|---|---|
| add Softargmax-L2 loss instead of KL loss (ablation) | CamCalib · Pano360 test | vfov error (deg): 3.53 -> 3.34 | Improved |
| add asymmetric biased loss for vfov (ablation) | CamCalib · Pano360 test | vfov error (deg): 3.34 -> 3.24 | Improved |
| add camera rotation Rc in projection (ablation) | HMR* · SPEC-SYN | W-MPJPE (mm): 118.3 / 85.1 -> 77.2 / 77.2 | Improved |
| add camera conditioning of the regressor (ablation) | HMR* + c + f + Rc · SPEC-SYN | W-MPJPE (mm): 77.2 -> 74.9 | Improved |
| use image center as principal point (ablation) | HMR* · SPEC-SYN | W-MPJPE (mm): 128.7 / 96.4 -> 120.4 / 84.2 | Improved |
| add CamCalib focal length instead of 5000 (ablation) | HMR* + c · SPEC-SYN | W-MPJPE (mm): 120.4 / 84.2 -> 118.3 / 85.1 | Mixed |
| add camera conditioning (ablation) | HMR* + c + f + Rc · 3DPW val | W-MPJPE (mm): 102.7 -> 98.1 | Improved |
| replace HMR* with SPEC | HMR* · 3DPW test | MPJPE (mm): 92.5 -> 96.5 | Hurt |
| replace HMR* with SPEC | HMR* · SPEC-SYN | W-MPJPE (mm): 128.7 / 96.4 -> 74.9 / 74.9 | Improved |
| add estimated rotation Rc to SMPLify-X (ablation) | SMPLify-X (CamCalib K) · SPEC-SYN | W-MPJPE (mm): 136.5 / 116.4 -> 115.2 / 115.2 | Improved |
| use CamCalib K instead of f=5000 in SMPLify-X (ablation) | SMPLify-X · SPEC-SYN | W-MPJPE (mm): 168.9 / 149.5 -> 136.5 / 116.4 | Improved |
| use CamCalib K and Rc instead of f=5000 in SMPLify-X | SMPLify-X · 3DPW val | W-MPJPE (mm): 123.8 / 94.7 -> 95.5 / 95.5 | Mixed |
Later papers that compared with this method
2 later accepted papers report SPEC as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- Depth Pro: Sharp Monocular Metric Depth in Less Than a Second (ICLR 2025 · better on 5 of 5 shared results)
- Multi-Person 3D Pose and Shape Estimation via Inverse Kinematics and Refinement (ECCV 2022 · better on 2 of 2 shared results)
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.
- SPIN — SPIN: Structure-Preserving Inner Offset Network for Scene Text Recognition (AAAI 2021) · better on 3 of 3 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
- Recovering Accurate 3D Human Pose in The Wild Using IMUs and a Moving Camera (ECCV 2018 · similarity 0.90)
- PointHPS: Cascaded 3D Human Pose and Shape Estimation from Point Clouds (IJCV 2026 · similarity 0.88)
- Benchmarking and Analyzing 3D Human Pose and Shape Estimation Beyond Algorithms (NeurIPS 2022 · similarity 0.88)
- ODHSR: Online Dense 3D Reconstruction of Humans and Scenes from Monocular Videos (CVPR 2025 · similarity 0.88)
- In the Wild Human Pose Estimation Using Explicit 2D Features and Intermediate 3D Representations (CVPR 2019 · similarity 0.88)
- Reconstructing People, Places, and Cameras (CVPR 2025 · similarity 0.88)
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.
