Add files using upload-large-folder tool

README.md

@@ -0,0 +1,84 @@
+# Cloud4D Dataset
+Dataset for "Cloud4D: Estimating Cloud Properties at a High Spatial and Temporal Resolution"
+Project Page: https://cloud4d.jacob-lin.com/
+
+## Dataset Structure
+
+The dataset is organized into two main categories:
+
+```
+Cloud4D/
+├── real_world/                    # Real-world stereo camera captures
+│   ├── 20230705_10/               # Date-hour folders (YYYYMMDD_HH)
+│   │   ├── perspective_1/
+│   │   │   ├── left_images/       # Left camera images
+│   │   │   ├── right_images/      # Right camera images
+│   │   │   ├── camera_pair.npz    # Stereo calibration data
+│   │   │   └── *.npy              # Camera intrinsics/extrinsics
+│   │   ├── perspective_2/
+│   │   └── perspective_3/
+│   └── ...
+└── synthetic/                     # Synthetic cloud renders
+    ├── terragen/                  # Terragen renders
+    │   ├── perspective_1/
+    │   ├── perspective_2/
+    │   └── perspective_3/
+    └── large_eddy_simulations/    # LES-based renders
+        ├── perspective_1/
+        ├── perspective_2/
+        ├── perspective_3/
+        └── volumes/
+```
+
+## Quick Start
+
+### Download and Extract
+
+```bash
+# Clone the dataset
+git clone https://huggingface.co/datasets/jacoblin/Cloud4D
+cd Cloud4D
+
+# Extract all archives
+python unpack.py
+
+# Or extract to a specific location
+python unpack.py --output /path/to/Cloud4D
+```
+
+### Selective Extraction
+
+```python
+# Extract only real-world data
+python unpack.py --subset real_world
+
+# Extract only synthetic data
+python unpack.py --subset synthetic
+
+# Extract a specific date
+python unpack.py --filter 20230705
+
+# List available archives without extracting
+python unpack.py --list
+```
+
+### Parallel Extraction (Faster)
+
+```bash
+# Use 4 parallel workers
+python unpack.py --jobs 4
+```
+
+## Citation
+
+If you use this dataset in your research, please cite:
+
+```bibtex
+@inproceedings{
+      lin2025cloudd,
+      title={Cloud4D: Estimating Cloud Properties at a High Spatial and Temporal Resolution},
+      author={Jacob Lin and Edward Gryspeerdt and Ronald Clark},
+      booktitle={The Thirty-ninth Annual Conference on Neural Information Processing Systems},
+      year={2025},
+      url={https://openreview.net/forum?id=g2AAvmBwkS}
+```

real_world/20230705_10.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8e5cea8bd8a34d7aacf29ca84a7d4906bd65e16e91b3f1e3cf174a34aac8858
+size 1161118885

real_world/20230705_11.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:56e5946d201708eb20d5c2ae8718bc3df172553f49c36542322ebf3fa0716ac8
+size 1203617465

real_world/20230717_10.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ecfb3e9296140e200449bc7e0e2367065c2a623ba01bb712b498d7eb1adec75d
+size 1295947783

real_world/20230717_11.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d047c515131ef530b24b4f84862783361b2ee1d6d4f0827f4549981d068e6436
+size 1224286008

real_world/20230720_10.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fddf681e62e53cb9bcf07e9acd0df47190aaa2ad07269b7c3cfb275489d56673
+size 1211943652

real_world/20230723_11.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:92262c4c50dc39cd33c3c529caa755d2dccb0c52e2aeaff725efb4040cfbc930
+size 758659974

real_world/20230725_10.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8950d99a35e88df10068f263f729214069e40a1b231a7539ea780f694663a724
+size 1062260243

real_world/20230806_11.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04109e242e497f7feb4bac560bbd733788f16943a96e5484101623fe9f958199
+size 1136773444

real_world/20230807_10.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:22ddaf0c9134ade34017de097166b6a564e35c330454feb9af7a856612516dbe
+size 1050388450

real_world/20230807_11.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2ec41228e42254cf6b11f6a1a36b5086214740453ff720bf90a8861bc6c4ab2
+size 1002149566

real_world/20230815_11.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b7306053c3570ce6c2f4ce6ccb83f6b8aa8610f2d2f32c3f1eb8befd7622c3c
+size 1201651863

real_world/20230821_10.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4cc096e7a53f833e0fb47c5e3e72bbc58fdcf3d75b99c2e6e24716e0baa40127
+size 1000207557

real_world/20230825_11.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b8bb0c767bf5250f01f8260c3a6ed8f18a16ab16a43b7da5c7bb836709a9c12
+size 1199614612

real_world/20230911_10.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fd6bdb2ec6bacb4b15ffb0a06b30f992b1b00667a0e7dfa19f6c0fc226353f1
+size 883999594

real_world/20230923_10.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc99bdbc5c797a29fbcf69687fc041cde521c51a95988d5ba06a60126832a4c4
+size 749512676

real_world/20230923_11.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ffc11b87098b982c8455a88e9ae47d413091610e195762f93c5d0c6d06a662a
+size 1138697915

real_world/20230923_12.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:71f541699ed45db6cfd85087de9f2e1e36147ab2dc7d3605fd3462f2b1870194
+size 1032600472

synthetic/large_eddy_simulations.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e205dc463d49f32a2b2b9d9ea98e4f8c5499e4b1c7269cd370aaed0afc820a51
+size 33098775283

synthetic/terragen.tar.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5343143787fd60933cc5b6add614d6d6a73af5d74fd764974f9b62c6043965b
+size 1641841189

unpack.py

@@ -0,0 +1,221 @@
+#!/usr/bin/env python3
+"""
+Unpack Cloud4D dataset archives.
+
+This script extracts all tar.gz archives to reconstruct the original
+Cloud4D directory structure.
+
+Usage:
+    python unpack.py [--output /path/to/output] [--subset real_world|synthetic] [--jobs N]
+
+Examples:
+    # Extract everything to ./Cloud4D
+    python unpack.py
+
+    # Extract to a specific location
+    python unpack.py --output /data/Cloud4D
+
+    # Extract only real_world data
+    python unpack.py --subset real_world
+
+    # Extract only synthetic data
+    python unpack.py --subset synthetic
+
+    # Extract a specific date-hour (real_world)
+    python unpack.py --filter 20230705_10
+
+    # Use parallel extraction
+    python unpack.py --jobs 4
+"""
+
+import argparse
+import subprocess
+import sys
+import tarfile
+from pathlib import Path
+from concurrent.futures import ProcessPoolExecutor, as_completed
+
+
+def extract_archive(archive_path, output_dir):
+    """Extract a tar.gz archive to the output directory."""
+    archive_path = Path(archive_path)
+    output_dir = Path(output_dir)
+
+    # Try using pigz for parallel decompression (faster)
+    try:
+        subprocess.run(['which', 'pigz'], check=True, capture_output=True)
+        cmd = f'pigz -dc "{archive_path}" | tar -xf - -C "{output_dir}"'
+        subprocess.run(cmd, shell=True, check=True)
+        return True
+    except (subprocess.CalledProcessError, FileNotFoundError):
+        pass
+
+    # Fallback to regular tar
+    try:
+        cmd = ['tar', '-xzf', str(archive_path), '-C', str(output_dir)]
+        subprocess.run(cmd, check=True)
+        return True
+    except subprocess.CalledProcessError:
+        pass
+
+    # Final fallback to Python tarfile
+    try:
+        with tarfile.open(archive_path, 'r:gz') as tar:
+            tar.extractall(output_dir)
+        return True
+    except Exception as e:
+        print(f"Error extracting {archive_path}: {e}")
+        return False
+
+
+def extract_single(args):
+    """Worker function for parallel extraction."""
+    archive_path, output_dir, name = args
+    try:
+        success = extract_archive(archive_path, output_dir)
+        if success:
+            return (name, 'extracted')
+        else:
+            return (name, 'failed')
+    except Exception as e:
+        return (name, f'error: {e}')
+
+
+def main():
+    parser = argparse.ArgumentParser(
+        description='Unpack Cloud4D dataset archives',
+        formatter_class=argparse.RawDescriptionHelpFormatter,
+        epilog="""
+Examples:
+    python unpack.py                           # Extract all to ./Cloud4D
+    python unpack.py --output /data/Cloud4D    # Extract to specific location
+    python unpack.py --subset real_world       # Extract only real_world
+    python unpack.py --filter 20230705         # Extract matching archives
+    python unpack.py --jobs 4                  # Parallel extraction
+        """
+    )
+    parser.add_argument('--output', '-o', type=Path, default=Path('./Cloud4D'),
+                        help='Output directory (default: ./Cloud4D)')
+    parser.add_argument('--subset', choices=['real_world', 'synthetic'],
+                        help='Extract only a specific subset')
+    parser.add_argument('--filter', type=str,
+                        help='Filter archives by name (e.g., "20230705" for a specific date)')
+    parser.add_argument('--jobs', '-j', type=int, default=1,
+                        help='Number of parallel extraction jobs (default: 1)')
+    parser.add_argument('--list', '-l', action='store_true',
+                        help='List available archives without extracting')
+    args = parser.parse_args()
+
+    # Find the script directory (where archives are located)
+    script_dir = Path(__file__).parent.resolve()
+
+    # Collect archives
+    archives = []
+
+    # Real world archives
+    real_world_dir = script_dir / 'real_world'
+    if real_world_dir.exists() and args.subset in (None, 'real_world'):
+        for archive in sorted(real_world_dir.glob('*.tar.gz')):
+            if args.filter is None or args.filter in archive.name:
+                archives.append(('real_world', archive))
+
+    # Synthetic archives
+    synthetic_dir = script_dir / 'synthetic'
+    if synthetic_dir.exists() and args.subset in (None, 'synthetic'):
+        for archive in sorted(synthetic_dir.glob('*.tar.gz')):
+            if args.filter is None or args.filter in archive.name:
+                archives.append(('synthetic', archive))
+
+    if not archives:
+        print("No archives found matching the criteria.")
+        print(f"Searched in: {script_dir}")
+        sys.exit(1)
+
+    # List mode
+    if args.list:
+        print("Available archives:")
+        print()
+        current_subset = None
+        for subset, archive in archives:
+            if subset != current_subset:
+                print(f"  {subset}/")
+                current_subset = subset
+            size_mb = archive.stat().st_size / 1024 / 1024
+            print(f"    {archive.name} ({size_mb:.1f} MB)")
+        print()
+        total_size = sum(a.stat().st_size for _, a in archives) / 1024 / 1024 / 1024
+        print(f"Total: {len(archives)} archives, {total_size:.2f} GB")
+        return
+
+    # Extract mode
+    output_dir = args.output.resolve()
+
+    print("=" * 70)
+    print("Cloud4D Dataset Unpacker")
+    print("=" * 70)
+    print(f"Output directory: {output_dir}")
+    print(f"Archives to extract: {len(archives)}")
+    if args.subset:
+        print(f"Subset: {args.subset}")
+    if args.filter:
+        print(f"Filter: {args.filter}")
+    print()
+
+    # Create output structure
+    (output_dir / 'real_world').mkdir(parents=True, exist_ok=True)
+    (output_dir / 'synthetic').mkdir(parents=True, exist_ok=True)
+
+    # Prepare extraction tasks
+    tasks = []
+    for subset, archive in archives:
+        target_dir = output_dir / subset
+        name = f"{subset}/{archive.stem}"
+        tasks.append((archive, target_dir, name))
+
+    # Extract
+    print("Extracting archives...")
+    results = []
+
+    if args.jobs > 1:
+        with ProcessPoolExecutor(max_workers=args.jobs) as executor:
+            futures = {executor.submit(extract_single, task): task[2] for task in tasks}
+            for future in as_completed(futures):
+                name, status = future.result()
+                results.append((name, status))
+                print(f"  [{status.upper()}] {name}")
+    else:
+        for task in tasks:
+            name, status = extract_single(task)
+            results.append((name, status))
+            print(f"  [{status.upper()}] {name}")
+
+    # Summary
+    print()
+    print("=" * 70)
+    print("EXTRACTION COMPLETE")
+    print("=" * 70)
+
+    extracted = sum(1 for _, s in results if s == 'extracted')
+    failed = sum(1 for _, s in results if s != 'extracted')
+
+    print(f"Successfully extracted: {extracted}")
+    if failed:
+        print(f"Failed: {failed}")
+    print()
+    print(f"Dataset extracted to: {output_dir}")
+    print()
+    print("Directory structure:")
+    print(f"  {output_dir}/")
+    print(f"    real_world/")
+    print(f"      20230705_10/")
+    print(f"        perspective_1/")
+    print(f"        perspective_2/")
+    print(f"        perspective_3/")
+    print(f"      ... (more date-hour folders)")
+    print(f"    synthetic/")
+    print(f"      terragen/")
+    print(f"      large_eddy_simulations/")
+
+
+if __name__ == '__main__':
+    main()