Document code utilities
Browse files
README.md
CHANGED
|
@@ -70,6 +70,7 @@ SP-TransientBench/
|
|
| 70 |
|-- Reconstruction/ multi-view reconstruction scene archives
|
| 71 |
|-- Annotations/ semantic annotation archives
|
| 72 |
|-- Histgram/ semantic-track raw histograms and depth maps
|
|
|
|
| 73 |
`-- config/ SPL calibration, extrinsics, parsing settings, and IRF data
|
| 74 |
```
|
| 75 |
|
|
@@ -150,6 +151,16 @@ Histgram/
|
|
| 150 |
|
| 151 |
Inside each sequence archive, semantic labels are stored as `.npy` arrays named like `RawDataHistogramMap_frame_*_semantic.npy`. Each array has shape `(49152, 672)`, where `49152 = 256 x 192`, and stores `uint8` semantic ids with `0` for unlabeled/background bins and `1-13` for semantic classes.
|
| 152 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 153 |
## Task Details
|
| 154 |
|
| 155 |
### Depth Estimation
|
|
@@ -199,6 +210,7 @@ The annotation pipeline identifies dominant peaks, assigns semantic labels to pe
|
|
| 199 |
| `Reconstruction/` | Scene packages for sparse-view SPL reconstruction and novel-view evaluation |
|
| 200 |
| `Annotations/` | Histogram-domain semantic label packages for `27` sequences |
|
| 201 |
| `Histgram/` | Raw histogram packages and sequence-level depth maps for the semantic track |
|
|
|
|
| 202 |
| `config/config.yaml` | SPL intrinsics, distortion coefficients, temporal bin width, image size, and parsing settings |
|
| 203 |
| `config/IRF_global.csv` | Global instrument response function (IRF) measurements for transient calibration and analysis |
|
| 204 |
| `config/final_extrinsic_sp_to_livox.txt` | SPL-to-Livox extrinsic calibration used for cross-sensor alignment |
|
|
|
|
| 70 |
|-- Reconstruction/ multi-view reconstruction scene archives
|
| 71 |
|-- Annotations/ semantic annotation archives
|
| 72 |
|-- Histgram/ semantic-track raw histograms and depth maps
|
| 73 |
+
|-- codes/ annotation, point-cloud conversion, simulation, and reconstruction tools
|
| 74 |
`-- config/ SPL calibration, extrinsics, parsing settings, and IRF data
|
| 75 |
```
|
| 76 |
|
|
|
|
| 151 |
|
| 152 |
Inside each sequence archive, semantic labels are stored as `.npy` arrays named like `RawDataHistogramMap_frame_*_semantic.npy`. Each array has shape `(49152, 672)`, where `49152 = 256 x 192`, and stores `uint8` semantic ids with `0` for unlabeled/background bins and `1-13` for semantic classes.
|
| 153 |
|
| 154 |
+
## Code
|
| 155 |
+
|
| 156 |
+
The `codes/` folder contains utility code released with STB:
|
| 157 |
+
|
| 158 |
+
- `codes/annotation tool/labelLidarwave.py`: a single-photon LiDAR waveform annotation tool.
|
| 159 |
+
- `codes/pointcloud_extract/raw2pc.py`: converts raw histogram `.txt` files into point clouds using the maximum-bin histogram method.
|
| 160 |
+
- `codes/pointcloud_extract/ann2pc.py`: converts annotated `.npy` semantic data into corresponding semantic point clouds.
|
| 161 |
+
- `codes/simulator/`: simulation tools used for STB experiments.
|
| 162 |
+
- `codes/reconstruction/`: STB reconstruction code based on the TransientNeRF method.
|
| 163 |
+
|
| 164 |
## Task Details
|
| 165 |
|
| 166 |
### Depth Estimation
|
|
|
|
| 210 |
| `Reconstruction/` | Scene packages for sparse-view SPL reconstruction and novel-view evaluation |
|
| 211 |
| `Annotations/` | Histogram-domain semantic label packages for `27` sequences |
|
| 212 |
| `Histgram/` | Raw histogram packages and sequence-level depth maps for the semantic track |
|
| 213 |
+
| `codes/` | Annotation, point-cloud conversion, simulation, and TransientNeRF-based reconstruction code |
|
| 214 |
| `config/config.yaml` | SPL intrinsics, distortion coefficients, temporal bin width, image size, and parsing settings |
|
| 215 |
| `config/IRF_global.csv` | Global instrument response function (IRF) measurements for transient calibration and analysis |
|
| 216 |
| `config/final_extrinsic_sp_to_livox.txt` | SPL-to-Livox extrinsic calibration used for cross-sensor alignment |
|