Research map / 3D vision / Human pose estimation
Ray3D: Ray-Based 3D Human Pose Estimation for Monocular Absolute 3D Localization
Yu Zhan, Fenghai Li, Renliang Weng, Wongun Choi
Check 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 normalized 3D rays and camera embedding instead of the RIE baseline | Absolute 3D pose estimation · Human3.6M test with CPN-detected 2D keypoints | Abs-MPJPE: 138.4 -> 109.5 | Improved |
| Use normalized 3D rays and camera embedding instead of the RIE baseline | Absolute 3D pose estimation · Human3.6M test with CPN-detected 2D keypoints | MRPE: 129.6 -> 99.6 | Improved |
| Use Ray3D instead of PoseFormer | Absolute 3D pose estimation · 3DHP test with GT 2D keypoints | Abs-MPJPE: 147.8 -> 103.4 | Improved |
| Use Ray3D instead of PoseFormer | Absolute 3D pose estimation · 3DHP test with GT 2D keypoints | MRPE: 147.5 -> 95.8 | Improved |
| Add temporal input using 9 frames instead of a single frame | Ray3D · Human3.6M test with CPN-detected 2D keypoints | Abs-MPJPE: 115.1 -> 109.5 | Mixed |
| Add intrinsic decoupling (ablation) | RIE · Human3.6M train → 3DHP test | MRPE: 1079.2 -> 448.9 | Improved |
| Add camera normalization after intrinsic decoupling (ablation) | RIE with intrinsic decoupling · Human3.6M train → 3DHP test | MRPE: 448.9 -> 311.6 | Improved |
| Add camera embedding (ablation) | Ray3D without camera embedding · Human3.6M train → 3DHP test | MRPE: 311.6 -> 307.4 | Improved |
| Use Ray3D instead of RIE | Root-relative pose estimation · Human3.6M test with GT 2D keypoints | MPJPE: 34.8 -> 34.4 | Improved |
| Convert pixel keypoints to normalized rays | Ray3D · Synthetic focal-length variation | MRPE | Improved |
| Add camera embedding to normalized-ray input (ablation) | Ray3D without camera embedding · Synthetic camera extrinsic variation | MRPE | Improved |
| Use noisy camera pitch as input | Ray3D · Human3.6M noisy-camera evaluation | MPJPE: 81.2mm to 110.6mm | Hurt |
Later papers that compared with this method
1 later accepted paper reports Ray3D as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- PoseAnchor: Robust Root Position Estimation for 3D Human Pose Estimation (ICCV 2025 · better on 2 of 2 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
- Consensus-Based Optimization for 3D Human Pose Estimation in Camera Coordinates (IJCV 2022 · similarity 0.89)
- Cameras as Rays: Pose Estimation via Ray Diffusion (ICLR 2024 · similarity 0.87)
- UniK3D: Universal Camera Monocular 3D Estimation (CVPR 2025 · similarity 0.87)
- A Simple yet Effective Baseline for 3D Human Pose Estimation (ICCV 2017 · similarity 0.87)
- Probabilistic Monocular 3D Human Pose Estimation With Normalizing Flows (ICCV 2021 · similarity 0.87)
- Neural Localizer Fields for Continuous 3D Human Pose and Shape Estimation (NeurIPS 2024 · 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.
