Gardado en:
| Main Authors: | , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Subjects: | |
| Acceso en liña: | https://arxiv.org/abs/2407.05921 |
| Tags: |
Engadir etiqueta
Sen Etiquetas, Sexa o primeiro en etiquetar este rexistro!
|
| _version_ | 1866913482077634560 |
|---|---|
| author | Koppula, Skanda Rocco, Ignacio Yang, Yi Heyward, Joe Carreira, João Zisserman, Andrew Brostow, Gabriel Doersch, Carl |
| author_facet | Koppula, Skanda Rocco, Ignacio Yang, Yi Heyward, Joe Carreira, João Zisserman, Andrew Brostow, Gabriel Doersch, Carl |
| contents | We introduce a new benchmark, TAPVid-3D, for evaluating the task of long-range Tracking Any Point in 3D (TAP-3D). While point tracking in two dimensions (TAP) has many benchmarks measuring performance on real-world videos, such as TAPVid-DAVIS, three-dimensional point tracking has none. To this end, leveraging existing footage, we build a new benchmark for 3D point tracking featuring 4,000+ real-world videos, composed of three different data sources spanning a variety of object types, motion patterns, and indoor and outdoor environments. To measure performance on the TAP-3D task, we formulate a collection of metrics that extend the Jaccard-based metric used in TAP to handle the complexities of ambiguous depth scales across models, occlusions, and multi-track spatio-temporal smoothness. We manually verify a large sample of trajectories to ensure correct video annotations, and assess the current state of the TAP-3D task by constructing competitive baselines using existing tracking models. We anticipate this benchmark will serve as a guidepost to improve our ability to understand precise 3D motion and surface deformation from monocular video. Code for dataset download, generation, and model evaluation is available at https://tapvid3d.github.io |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_05921 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | TAPVid-3D: A Benchmark for Tracking Any Point in 3D Koppula, Skanda Rocco, Ignacio Yang, Yi Heyward, Joe Carreira, João Zisserman, Andrew Brostow, Gabriel Doersch, Carl Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning We introduce a new benchmark, TAPVid-3D, for evaluating the task of long-range Tracking Any Point in 3D (TAP-3D). While point tracking in two dimensions (TAP) has many benchmarks measuring performance on real-world videos, such as TAPVid-DAVIS, three-dimensional point tracking has none. To this end, leveraging existing footage, we build a new benchmark for 3D point tracking featuring 4,000+ real-world videos, composed of three different data sources spanning a variety of object types, motion patterns, and indoor and outdoor environments. To measure performance on the TAP-3D task, we formulate a collection of metrics that extend the Jaccard-based metric used in TAP to handle the complexities of ambiguous depth scales across models, occlusions, and multi-track spatio-temporal smoothness. We manually verify a large sample of trajectories to ensure correct video annotations, and assess the current state of the TAP-3D task by constructing competitive baselines using existing tracking models. We anticipate this benchmark will serve as a guidepost to improve our ability to understand precise 3D motion and surface deformation from monocular video. Code for dataset download, generation, and model evaluation is available at https://tapvid3d.github.io |
| title | TAPVid-3D: A Benchmark for Tracking Any Point in 3D |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2407.05921 |