Research map / 3D vision / Human pose estimation
FSA-Net: Learning Fine-Grained Structure Aggregation for Head Pose Estimation From a Single Image
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 fine-grained structure mapping and capsule feature aggregation | SSR-Net-MD · AFLW2000, protocol 1 | MAE: 6.01 -> 5.07 | Helped |
| add fine-grained structure mapping and capsule feature aggregation | SSR-Net-MD · BIWI, protocol 1 | MAE: 4.65 -> 4.00 | Helped |
| use a single RGB frame instead of RGB plus time | Head pose estimation on BIWI protocol 2 · BIWI, protocol 2 | yaw MAE: 3.14 -> 2.89 | Helped |
| use a single RGB frame instead of RGB plus time | Head pose estimation on BIWI protocol 2 · BIWI, protocol 2 | MAE: 3.07 -> 3.60 | Hurt |
| add fine-grained structure mapping (ablation) | FSA-Net with capsule aggregation · AFLW2000, protocol 1 | MAE: 5.75 -> 5.25 | Helped |
| add fine-grained structure mapping (ablation) | FSA-Net with capsule aggregation · BIWI, protocol 1 | MAE: 4.22 -> 4.31 | Hurt |
| use variance scoring instead of uniform scoring (ablation) | FSA-Net with capsule aggregation · AFLW2000, protocol 1 | MAE: 5.75 -> 5.36 | Helped |
| use variance scoring instead of uniform scoring (ablation) | FSA-Net with capsule aggregation · BIWI, protocol 1 | MAE: 4.22 -> 4.28 | Hurt |
| average predictions from the 1×1, variance, and uniform-scoring variants (ablation) | FSA-Net with capsule aggregation · BIWI, protocol 1 | MAE: 4.31 -> 4.00 | Helped |
| add capsule feature aggregation (ablation) | FSA-Net without feature aggregation · AFLW2000, protocol 1 | MAE: 5.54 -> 5.25 | Helped |
| add capsule feature aggregation (ablation) | FSA-Net without feature aggregation · BIWI, protocol 1 | MAE: 4.61 -> 4.31 | Helped |
| add capsule feature aggregation (ablation) | FSA-Net without feature aggregation · BIWI, protocol 2 | MAE: 3.95 -> 3.77 | Helped |
Closest papers by meaning
- Video-Based Human Pose Regression via Decoupled Space-Time Aggregation (CVPR 2024 · similarity 0.84)
- GIVEPose: Gradual Intra-class Variation Elimination for RGB-based Category-Level Object Pose Estimation (CVPR 2025 · similarity 0.84)
- FinePOSE: Fine-Grained Prompt-Driven 3D Human Pose Estimation via Diffusion Models (CVPR 2024 · similarity 0.83)
- Recurrent Feature Mining and Keypoint Mixup Padding for Category-Agnostic Pose Estimation (CVPR 2025 · similarity 0.83)
- UNOPose: Unseen Object Pose Estimation with an Unposed RGB-D Reference Image (CVPR 2025 · similarity 0.82)
- CAP-Net: A Unified Network for 6D Pose and Size Estimation of Categorical Articulated Parts from a Single RGB-D Image (CVPR 2025 · similarity 0.82)
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.