Research map / 3D vision / Human pose estimation
Sub-Frame Appearance and 6D Pose Estimation of Fast Moving Objects
Denys Rozumnyi, Jan Kotera, Filip Sroubek, Jiri Matas
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 |
|---|---|---|---|
| Extend 2D tracking with piece-wise deblatting and 3D trajectory estimation | TbD-3D · TbD-3D dataset, 30 fps | TIoU-3D: .500 -> .539 | Improved |
| Extend 2D tracking with piece-wise deblatting and 3D trajectory estimation | TbD-3D · TbD-3D dataset, 30 fps | TIoU-3D: .447 -> .539 | Improved |
| Estimate object size and depth from the reconstructed shape | TbD-3D · TbD-3D dataset, 30 fps | Radius error [pixels]: 5.157 -> 3.122 | Improved |
| Use the ground-truth 2D trajectory instead of an estimated 2D trajectory | TbD-3D · TbD-3D dataset, 30 fps | TIoU-3D: .539 -> .924 | Improved |
| Use the ground-truth 2D trajectory instead of an estimated 2D trajectory | TbD-3D · TbD-3D dataset, 30 fps | Radius error [pixels]: 3.122 -> 1.492 | Improved |
| Use the ground-truth 2D trajectory instead of an estimated 2D trajectory | TbD-3D · TbD-3D dataset, 30 fps | Axis error [degrees]: 38.857 -> 37.195 | Improved |
| Use the ground-truth 2D trajectory instead of an estimated 2D trajectory | TbD-3D · TbD-3D dataset, 30 fps | Angle error [degrees]: 2.467 -> 2.250 | Improved |
| Estimate 3D rotation axis and angle from reconstructed appearance snapshots | TbD-3D · Rolling-ball sequences with known trajectory length | Axis error [degrees]: 4.052 degrees | Improved |
| Estimate rotation angle from reconstructed appearance snapshots | TbD-3D · Rolling-ball sequences with known trajectory length | Relative rotation-angle error: 0.037 degrees; 1.2 % | Improved |
Closest papers by meaning
- Tracking by Deblatting (IJCV 2021 · similarity 0.88)
- Motion-From-Blur: 3D Shape and Motion Estimation of Motion-Blurred Objects in Videos (CVPR 2022 · similarity 0.88)
- MonST3R: A Simple Approach for Estimating Geometry in the Presence of Motion (ICLR 2025 · similarity 0.85)
- DELTA: DENSE EFFICIENT LONG-RANGE 3D TRACKING FOR ANY VIDEO (ICLR 2025 · similarity 0.84)
- SpatialTracker: Tracking Any 2D Pixels in 3D Space (CVPR 2024 · similarity 0.84)
- 6D Object Pose Tracking in Internet Videos for Robotic Manipulation (ICLR 2025 · similarity 0.84)
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.
