Update README.md
Browse files
README.md
CHANGED
@@ -17,17 +17,17 @@ However, the *complete dataset generating process* is provided as followed:
|
|
17 |
1. Clone the official MMDetection3D repository.
|
18 |
|
19 |
```bash
|
20 |
-
git clone https://github.com/open-mmlab/mmdetection3d.git
|
21 |
```
|
22 |
|
23 |
2. Swith to `v1.3.0` version.
|
24 |
|
25 |
```bash
|
26 |
-
cd
|
27 |
git checkout v1.3.0
|
28 |
```
|
29 |
|
30 |
-
3. Download the *ScanNet* dataset following [instructions](https://github.com/open-mmlab/mmdetection3d/tree/v1.3.0/data/scannet) and place `scans` directory as `
|
31 |
|
32 |
> :bulb: Note
|
33 |
>
|
@@ -65,7 +65,7 @@ scannet
|
|
65 |
|
66 |
The ground truth labels we used are from [SCFusion](https://github.com/ShunChengWu/SCFusion#generate-gt). Ground truth is available at [here](https://github.com/ShunChengWu/SCFusion#generate-gt).
|
67 |
|
68 |
-
The ground truth label should be placed as `
|
69 |
|
70 |
6. Reformulate *CompleteScanNet*
|
71 |
|
@@ -73,7 +73,7 @@ The ground truth label should be placed as `ISO/data/completescannet/gt`.
|
|
73 |
python preprocess_gt.py
|
74 |
```
|
75 |
|
76 |
-
The resulted directory is `
|
77 |
|
78 |
Now, we obtained the following directory structure.
|
79 |
|
@@ -89,7 +89,7 @@ completescannet
|
|
89 |
|
90 |
7. Create the *OccScanNet*
|
91 |
|
92 |
-
First, you should create a directories with name `preprocessed_voxels` and `gathered_data` in data disk and link them to the `
|
93 |
|
94 |
```bash
|
95 |
python generate_gt.py
|
|
|
17 |
1. Clone the official MMDetection3D repository.
|
18 |
|
19 |
```bash
|
20 |
+
git clone https://github.com/open-mmlab/mmdetection3d.git ISO_mm
|
21 |
```
|
22 |
|
23 |
2. Swith to `v1.3.0` version.
|
24 |
|
25 |
```bash
|
26 |
+
cd ISO_mm
|
27 |
git checkout v1.3.0
|
28 |
```
|
29 |
|
30 |
+
3. Download the *ScanNet* dataset following [instructions](https://github.com/open-mmlab/mmdetection3d/tree/v1.3.0/data/scannet) and place `scans` directory as `ISO_mm/data/scannet/scans`.
|
31 |
|
32 |
> :bulb: Note
|
33 |
>
|
|
|
65 |
|
66 |
The ground truth labels we used are from [SCFusion](https://github.com/ShunChengWu/SCFusion#generate-gt). Ground truth is available at [here](https://github.com/ShunChengWu/SCFusion#generate-gt).
|
67 |
|
68 |
+
The ground truth label should be placed as `ISO_mm/data/completescannet/gt`.
|
69 |
|
70 |
6. Reformulate *CompleteScanNet*
|
71 |
|
|
|
73 |
python preprocess_gt.py
|
74 |
```
|
75 |
|
76 |
+
The resulted directory is `ISO_mm/data/completescannet/preprocessed`.
|
77 |
|
78 |
Now, we obtained the following directory structure.
|
79 |
|
|
|
89 |
|
90 |
7. Create the *OccScanNet*
|
91 |
|
92 |
+
First, you should create a directories with name `preprocessed_voxels` and `gathered_data` in data disk and link them to the `ISO_mm/data/occscannet`.
|
93 |
|
94 |
```bash
|
95 |
python generate_gt.py
|