Update README.md
Browse files
README.md
CHANGED
|
@@ -6,7 +6,11 @@ license: apache-2.0
|
|
| 6 |
|
| 7 |
Download [FlowOcc3D](https://huggingface.co/viewformer/ViewFormer-Occ/blob/main/occ_flow_sparse_ext.zip). Unzip it in `./data/nuscenes`.
|
| 8 |
|
| 9 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
```python
|
| 11 |
W, H, Z = 200, 200, 16
|
| 12 |
sample_idx = results['sample_idx'] # nuScenes sample token
|
|
|
|
| 6 |
|
| 7 |
Download [FlowOcc3D](https://huggingface.co/viewformer/ViewFormer-Occ/blob/main/occ_flow_sparse_ext.zip). Unzip it in `./data/nuscenes`.
|
| 8 |
|
| 9 |
+
Please follow instructions of [Occ3D](https://github.com/CVPR2023-3D-Occupancy-Prediction/CVPR2023-3D-Occupancy-Prediction) to download nuScenes subdataset.
|
| 10 |
+
|
| 11 |
+
Our FlowOcc3D assigns a flow vector to each foreground occupancy of Occ3D.
|
| 12 |
+
|
| 13 |
+
Here we briefly introduce what the downloaded FlowOcc3D data looks like. We store the flow and index of the foreground voxels in the `.bin` file and the `_idx.bin` file.
|
| 14 |
```python
|
| 15 |
W, H, Z = 200, 200, 16
|
| 16 |
sample_idx = results['sample_idx'] # nuScenes sample token
|