Research map / 3D vision / Human pose estimation
In the Wild Human Pose Estimation Using Explicit 2D Features and Intermediate 3D Representations
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 explicit 2D pose loss in latent feature space and outdoor 2D data (ablation) | Direct 3D pose baseline with bone loss · MPI-INF-3DHP | PCK: 62.3 -> 66.4 | Helped |
| Add 3D-to-2D projection loss and outdoor data without explicit 2D latent loss (ablation) | Direct 3D pose baseline with bone loss · MPI-INF-3DHP | PCK: 62.3 -> 69.5 | Helped |
| Add both explicit 2D latent loss and 3D-to-2D projection loss with outdoor data (ablation) | Direct 3D pose baseline with bone loss · MPI-INF-3DHP | PCK: 62.3 -> 70.4 | Helped |
| Use both explicit 2D latent loss and 3D-to-2D projection loss instead of either component alone (ablation) | 3D pose model with outdoor 2D training data · MPI-INF-3DHP | PCK: 66.4 and 69.5 -> 70.4 | Helped |
| Train on more H3.6M frames, sampling at 5 fps instead of using H80K samples | Proposed pose estimation method · H3.6M | MPJPE (glob. scaled): 71.1 -> 65.7 | Helped |
| Use the in-the-wild-oriented method on the studio-biased H3.6M test set | Proposed pose estimation method · H3.6M | MPJPE: 71.1 vs. 74.1 | Mixed |
Closest papers by meaning
- 3D Human Pose Estimation in the Wild by Adversarial Learning (CVPR 2018 · similarity 0.93)
- Weakly-Supervised 3D Human Pose Learning via Multi-View Images in the Wild (CVPR 2020 · similarity 0.91)
- CanonPose: Self-Supervised Monocular 3D Human Pose Estimation in the Wild (CVPR 2021 · similarity 0.90)
- 3D Human Pose Estimation = 2D Pose Estimation + Matching (CVPR 2017 · similarity 0.90)
- Learning Monocular 3D Human Pose Estimation From Multi-View Images (CVPR 2018 · similarity 0.90)
- MoCap-guided Data Augmentation for 3D Pose Estimation in the Wild (NeurIPS 2016 · similarity 0.90)
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.