# Sub-frame Appearance and 6D Pose Estimation of Fast Moving Objects

Source: [Sub-frame Appearance and 6D Pose Estimation of Fast Moving Objects](<https://www.sota2.com/research/paper/68fa689e1e209131cea1d202>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-04T03:11:33.342Z

Authors: Denys Rozumnyi, Jan Kotera, Filip Sroubek, Jiri Matas

Published: 2019-11-25T14:13:25.000Z

## Abstract

We propose a novel method that tracks fast moving objects, mainly non-uniform spherical, in full 6 degrees of freedom, estimating simultaneously their 3D motion trajectory, 3D pose and object appearance changes with a time step that is a fraction of the video frame exposure time. The sub-frame object localization and appearance estimation allows realistic temporal super-resolution and precise shape estimation. The method, called TbD-3D (Tracking by Deblatting in 3D) relies on a novel reconstruction algorithm which solves a piece-wise deblurring and matting problem. The 3D rotation is estimated by minimizing the reprojection error. As a second contribution, we present a new challenging dataset with fast moving objects that change their appearance and distance to the camera. High speed camera recordings with zero lag between frame exposures were used to generate videos with different frame rates annotated with ground-truth trajectory and pose.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [3D Object Location and Pose Estimation on TbD-3D 1.0 (Average)](<https://www.sota2.com/research/sota/3d-object-location-and-pose-estimation-on-tbd-3d-1-0-average.md>): 3D Object Location and Pose Estimation; TbD-3D 1.0 (Average); TIoU-3D: 93.7.

- [Fast Moving Object Deblurring and Trajectory Estimation on TbD v1 (test)](<https://www.sota2.com/research/sota/fast-moving-object-deblurring-and-trajectory-estimation-on-tbd-v1-test.md>): Fast Moving Object Deblurring and Trajectory Estimation; TbD v1 (test); PSNR: 26.36.

- [Fast Moving Object Deblurring and Trajectory Estimation on TbD-3D v1 (test)](<https://www.sota2.com/research/sota/fast-moving-object-deblurring-and-trajectory-estimation-on-tbd-3d-v1-test.md>): Fast Moving Object Deblurring and Trajectory Estimation; TbD-3D v1 (test); PSNR: 24.84.

- [Fast Moving Object Deblurring and Trajectory Estimation on Falling Objects v1 (test)](<https://www.sota2.com/research/sota/fast-moving-object-deblurring-and-trajectory-estimation-on-falling-objects-v1-test.md>): Fast Moving Object Deblurring and Trajectory Estimation; Falling Objects v1 (test); PSNR: 23.42.

- [Fast moving object deblurring on TbD Dataset](<https://www.sota2.com/research/sota/fast-moving-object-deblurring-on-tbd-dataset.md>): Fast moving object deblurring; TbD Dataset; PSNR: 25.21.

- [Fast moving object deblurring on Falling Objects](<https://www.sota2.com/research/sota/fast-moving-object-deblurring-on-falling-objects.md>): Fast moving object deblurring; Falling Objects; PSNR: 23.42.

- [Fast moving object deblurring on TbD-3D Dataset](<https://www.sota2.com/research/sota/fast-moving-object-deblurring-on-tbd-3d-dataset.md>): Fast moving object deblurring; TbD-3D Dataset; PSNR: 23.13.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/1911.10927>)

[PDF](<https://arxiv.org/pdf/1911.10927.pdf>)