update image in readme
Browse files- README.md +12 -13
- assets/ADT_intro.jpg +3 -0
README.md
CHANGED
|
@@ -3,9 +3,9 @@ license: other
|
|
| 3 |
license_name: project-aria-privacy-policy
|
| 4 |
license_link: https://www.projectaria.com/privacy-policy/
|
| 5 |
---
|
| 6 |
-
|
| 7 |
# Aria Digital Twin (ADT) Dataset
|
| 8 |
-
|
|
|
|
| 9 |
## Dataset Summary
|
| 10 |
|
| 11 |
ADT provides raw and synthesized sensor data from Project Aria glasses, combined with groundtruth data generated using a motion capture system including depth images, device trajectories, object trajectories and bounding boxes, and human tracking. We also provide processed sensor data from our Machine Perception Services. Go to ADT Data Format to see a full list of the data we provide.
|
|
@@ -18,17 +18,16 @@ Go to the projectaria.com/datasets/adt/ to learn more about the dataset, includi
|
|
| 18 |
|
| 19 |
ADT contains 236 sequences, each sequence consists of a single Aria device recording (in VRS format) along with all ground truth data that accompanies this recording. When multiple devices were recording data concurrently, they are always time synchronized and all data collected in each scene is expresssed in the same World coordinate frame. Along with ground truth data, we also provide output from our Machine Perception Services (MPS) which you can learn more about [here](https://facebookresearch.github.io/projectaria_tools/docs/ARK/mps). All data included in each sequences is broken down in the table below:
|
| 20 |
|
| 21 |
-
|
| 22 |
-
|
|
| 23 |
-
|
| 24 |
-
|
|
| 25 |
-
|
|
| 26 |
-
|
|
| 27 |
-
|
|
| 28 |
-
| MPS
|
| 29 |
-
| MPS SLAM
|
| 30 |
-
| MPS SLAM
|
| 31 |
-
| MPS SLAM online calibration | Online calibrations | < 20 MB | 1.2 GB |
|
| 32 |
|
| 33 |
## Citation Information
|
| 34 |
|
|
|
|
| 3 |
license_name: project-aria-privacy-policy
|
| 4 |
license_link: https://www.projectaria.com/privacy-policy/
|
| 5 |
---
|
|
|
|
| 6 |
# Aria Digital Twin (ADT) Dataset
|
| 7 |
+
|
| 8 |
+
Figure 1: An overview of the ADT dataset
|
| 9 |
## Dataset Summary
|
| 10 |
|
| 11 |
ADT provides raw and synthesized sensor data from Project Aria glasses, combined with groundtruth data generated using a motion capture system including depth images, device trajectories, object trajectories and bounding boxes, and human tracking. We also provide processed sensor data from our Machine Perception Services. Go to ADT Data Format to see a full list of the data we provide.
|
|
|
|
| 18 |
|
| 19 |
ADT contains 236 sequences, each sequence consists of a single Aria device recording (in VRS format) along with all ground truth data that accompanies this recording. When multiple devices were recording data concurrently, they are always time synchronized and all data collected in each scene is expresssed in the same World coordinate frame. Along with ground truth data, we also provide output from our Machine Perception Services (MPS) which you can learn more about [here](https://facebookresearch.github.io/projectaria_tools/docs/ARK/mps). All data included in each sequences is broken down in the table below:
|
| 20 |
|
| 21 |
+
| Data type | What's included | Per sequence size | Total size for all sequences |
|
| 22 |
+
| :-------------------------- | :--------------------------------------------------------------------------------------------- | :---------------- | :--------------------------- |
|
| 23 |
+
| main Aria | Aria raw data, 2D bounding box, 3D object poses and bounding box, skeleton data, eye gaze data | 3 - 6 GB | ~700 GB |
|
| 24 |
+
| segmentation | Instance segmentation data | 2 - 4 GB | ~750 GB |
|
| 25 |
+
| depth | Depth map data | 4 - 8 GB | ~1.5 TB |
|
| 26 |
+
| synthetic | Synthetic rendering data | 2 - 4 GB | 500 GB |
|
| 27 |
+
| MPS eyegaze | Eyegaze, summary file | < 1 MB | ~31 MB |
|
| 28 |
+
| MPS SLAM points | Semi-dense points and observations | 200 - 500 MB | ~31 GB |
|
| 29 |
+
| MPS SLAM trajectories | Open and closed loop trajectories | 100 - 200 MB | 12 GB |
|
| 30 |
+
| MPS SLAM online calibration | Online calibrations | < 20 MB | 1.2 GB |
|
|
|
|
| 31 |
|
| 32 |
## Citation Information
|
| 33 |
|
assets/ADT_intro.jpg
ADDED
|
Git LFS Details
|