Research map / 3D vision / Human pose estimation
CanonPose: Self-Supervised Monocular 3D Human Pose Estimation in the Wild
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 the static-camera consistency constraint | CanonPose · Human3.6M | MPJPE: 81.9 -> 74.3 | Helped |
| Add the static-camera consistency constraint | CanonPose · MPI-INF-3DHP | PCK: 69.0 -> 77.0 | Helped |
| Use only the first two or three cameras instead of all four (ablation) | CanonPose · Human3.6M | MPJPE: 81.9 -> 82.7; 81.9 -> 82.0 | Hurt |
| Add joint confidences to the input and reprojection loss (ablation) | CanonPose · Human3.6M | MPJPE: 95.6 -> 81.9; gain of 19.4mm in MPJPE | Helped |
| Add joint confidences to the input and reprojection loss (ablation) | CanonPose · Human3.6M | PMPJPE: 65.0 -> 53.0; gain of 11.2mm in PMPJPE | Helped |
| Use ground-truth 2D annotations instead of detected 2D poses (ablation) | CanonPose · Human3.6M | MPJPE: 81.9 -> 65.9 | Helped |
| Replace mixed-view reprojection consistency with direct pose equality (ablation) | CanonPose · Human3.6M | MPJPE: 81.9 -> 554.3 | Hurt |
| Replace mixed-view reprojection consistency with direct camera equality (ablation) | CanonPose · Human3.6M | MPJPE: 617.9 | Hurt |
Closest papers by meaning
- Learning Monocular 3D Human Pose Estimation From Multi-View Images (CVPR 2018 · similarity 0.92)
- Weakly-Supervised 3D Human Pose Learning via Multi-View Images in the Wild (CVPR 2020 · similarity 0.92)
- MetaPose: Fast 3D Pose From Multiple Views Without 3D Supervision (CVPR 2022 · similarity 0.91)
- Self-Supervised 3D Human Pose Estimation via Part Guided Novel Image Synthesis (CVPR 2020 · similarity 0.90)
- SelfPose3d: Self-Supervised Multi-Person Multi-View 3d Pose Estimation (CVPR 2024 · similarity 0.90)
- COMPOSE: Hypergraph Cover Optimization for Multi-view 3D Human Pose Estimation (NeurIPS 2026 · 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.