Upload folder using huggingface_hub

README.md

@@ -1,3 +1,80 @@
----
-license: mit
----
+# SynLiDAR Dataset
+
+## Overview
+
+SynLiDAR is a synthetic LiDAR dataset designed for autonomous driving research.
+It contains high-quality simulated point cloud sequences and corresponding semantic annotations.
+
+The dataset provides two variants:
+- FullDataset – complete version for large-scale experiments
+- SubDataset – smaller version suitable for prototyping, debugging, and benchmarking
+
+## Dataset Structure
+```
+SynLiDAR/  
+├── FullDataset/  
+│   ├── sequences/  
+│   │   ├── 00.zip  
+│   │   ├── ...  
+│   │   └── 12.zip  
+│   └── readme.txt  
+│  
+├── SubDataset/  
+│   ├── sequences/  
+│   │   ├── 00.zip  
+│   │   ├── ...  
+│   │   └── 12.zip  
+│   └── readme.txt  
+│  
+├── annotations.yaml  
+├── read_data.py  
+└── README.md  
+```
+
+## Contents
+- sequences/*.zip — Each zip contains LiDAR frames
+- annotations.yaml — Semantic categories and their mappings
+- read_data.py — Example Python loader
+- readme.txt — Additional notes from the original dataset
+
+## License
+The SynLiDAR dataset in this repository is released under the MIT License.
+```
+MIT License
+
+Copyright (c) 2025 Aoran XIAO
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+```
+
+⚠️ **Disclaimer**: This dataset is intended for research and educational usage.
+Make sure to respect local regulations when training or deploying autonomous driving systems.
+
+## Citation
+```
+@inproceedings{xiao2022transfer,  
+  title={Transfer learning from synthetic to real lidar point cloud for semantic segmentation},  
+  author={Xiao, Aoran and Huang, Jiaxing and Guan, Dayan and Zhan, Fangneng and Lu, Shijian},  
+  booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},  
+  volume={36},  
+  number={3},  
+  pages={2795--2803},  
+  year={2022}  
+}
+```

annotations.yaml

@@ -0,0 +1,157 @@
+# This file is covered by the LICENSE file in the root of this project.
+labels:
+  0 : "unlabeled"
+  1: "car"
+  2: "pick-up"
+  3: "truck"
+  4: "bus"
+  5: "bicycle"
+  6: "motorcycle"
+  7: "other-vehicle"
+  8: "road"
+  9: "sidewalk"
+  10: "parking"
+  11: "other-ground"
+  12: "female"
+  13: "male"
+  14: "kid"
+  15: "crowd"  # multiple person that are very close
+  16: "bicyclist"
+  17: "motorcyclist"
+  18: "building"
+  19: "other-structure"
+  20: "vegetation"
+  21: "trunk"
+  22: "terrain"
+  23: "traffic-sign"
+  24: "pole"
+  25: "traffic-cone"
+  26: "fence"
+  27: "garbage-can"
+  28: "electric-box"
+  29: "table"
+  30: "chair"
+  31: "bench"
+  32: "other-object"
+color_map: # bgr
+  0 : [0, 0, 0]
+  1 : [0, 0, 255]
+  2: [245, 150, 100]
+  3: [245, 230, 100]
+  4: [250, 80, 100]
+  5: [150, 60, 30]
+  6: [255, 0, 0]
+  7: [180, 30, 80]
+  8: [255, 0, 0]
+  9: [30, 30, 255]
+  10: [200, 40, 255]
+  11: [90, 30, 150]
+  12: [255, 0, 255]
+  13: [255, 150, 255]
+  14: [75, 0, 75]
+  15: [75, 0, 175]
+  16: [0, 200, 255]
+  17: [50, 120, 255]
+  18: [0, 150, 255]
+  19: [170, 255, 150]
+  20: [0, 175, 0]
+  21: [0, 60, 135]
+  22: [80, 240, 150]
+  23: [150, 240, 255]
+  24: [0, 0, 255]
+  25: [255, 255, 50]
+  26: [245, 150, 100]
+  27: [255, 0, 0]
+  28: [200, 40, 255]
+  29: [30, 30, 255]
+  30: [90, 30, 150]
+  31: [250, 80, 100]
+  32: [180, 30, 80]
+# An example of class mapping from synlidar to semantickitti,
+# classes that are indistinguishable from single scan or inconsistent in
+# ground truth are mapped to their closest equivalent.
+map_2_semantickitti:
+  0: 0  # "unlabeled"
+  1: 1  # "car"
+  2: 4  # "pick-up"
+  3: 4  # "truck"
+  4: 5  # "bus"
+  5: 2  # "bicycle"
+  6: 3  # "motorcycle"
+  7: 5  # "other-vehicle"
+  8: 9  # "road"
+  9: 11  # "sidewalk"
+  10: 10  # "parking"
+  11: 12  # "other-ground"
+  12: 6  # "female"
+  13: 6  # "male"
+  14: 6  # "kid"
+  15: 6  # "crowd"
+  16: 7  # "bicyclist"
+  17: 8  # "motorcyclist"
+  18: 13  # "building"
+  19: 0  # "other-structure"
+  20: 15  # "vegetation"
+  21: 16  # "trunk"
+  22: 17  # "terrain"
+  23: 19  # "traffic-sign"
+  24: 18  # "pole"
+  25: 0  # "traffic-cone"
+  26: 14  # "fence"
+  27: 0  # "garbage-can"
+  28: 0  # "electric-box"
+  29: 0  # "table"
+  30: 0  # "chair"
+  31: 0  # "bench"
+  32: 0  # "other-object"
+# An example of class mapping from synlidar to semanticposs,
+# classes that are indistinguishable from single scan or inconsistent in
+# ground truth are mapped to their closest equivalent.
+map_2_semanticposs:
+  0: 255 # "unlabeled"
+  1: 2   # "car"
+  2: 2  
+  3: 2
+  4: 2
+  5: 11  # "bike"
+  6: 11
+  7: 255
+  8: 12  # "ground"
+  9: 12
+  10: 12
+  11: 12
+  12: 0  # "person"
+  13: 0
+  14: 0
+  15: 0
+  16: 1  # "rider"
+  17: 1
+  18: 8  # "building"
+  19: 255
+  20: 4  # "plant"
+  21: 3  # "trunk"
+  22: 4
+  23: 5  # "traffic-sign"
+  24: 6  # "pole"
+  25: 9  # "cone/stone"
+  26: 10 # "fence"
+  27: 7  # "trashcan"
+  28: 255  
+  29: 255 
+  30: 255 
+  31: 255 
+  32: 255
+sequences: # sequence numbers
+  - 00
+  - 01
+  - 02
+  - 03
+  - 04
+  - 05
+  - 06
+  - 07
+  - 08
+  - 09
+  - 10
+  - 11
+  - 12

read_data.py

@@ -0,0 +1,23 @@
+import numpy as np
+import glob
+
+def read_points(path):
+  scan = np.fromfile(path, dtype=np.float32)
+  scan = scan.reshape((-1, 4))  # [x,y,z,intensity]
+  return scan
+
+def read_label(path):
+  label = np.fromfile(path, dtype=np.uint32)
+  label = label.reshape((-1))
+  return label
+
+if __name__ == '__main__':
+    files = glob.glob('./*/velodyne/*.bin')
+
+    for f_path in files:
+      scan = read_points(f_path)
+
+      label_path = f_path.replace('velodyne', 'labels').replace('bin', 'label')
+      labels = read_label(label_path)
+
+      assert scan.shape[0] == labels.shape[0]