The dataset viewer is not available for this subset.
Exception: SplitsNotFoundError
Message: The split names could not be parsed from the dataset config.
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.12/site-packages/datasets/inspect.py", line 286, in get_dataset_config_info
for split_generator in builder._split_generators(
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/webdataset/webdataset.py", line 83, in _split_generators
raise ValueError(
ValueError: The TAR archives of the dataset should be in WebDataset format, but the files in the archive don't share the same prefix or the same types.
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/split_names.py", line 66, in compute_split_names_from_streaming_response
for split in get_dataset_split_names(
^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/inspect.py", line 340, in get_dataset_split_names
info = get_dataset_config_info(
^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/inspect.py", line 291, in get_dataset_config_info
raise SplitsNotFoundError("The split names could not be parsed from the dataset config.") from err
datasets.inspect.SplitsNotFoundError: The split names could not be parsed from the dataset config.Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
Replica Subset (3DVLM)
A small, fast-to-download slice of the Replica indoor scans, converted to a
uniform posed-RGB-D format for quick model test-runs. This is a subset of the
full set: 100 frames per scene, all 8 scenes (frames uniformly sampled along
each trajectory). For the complete data, see the full (private) 3dvlm-replica.
Contents
8 scenes (office0–office4, room0–room2), one .tar each under replica/.
Each tar extracts to a scene directory:
office0/
├── images/ # frame_000000.jpg … (100 RGB frames, 680×1200)
├── depth.npy # (100, 680, 1200) float32
├── valid_mask.npy # (100, 680, 1200) bool — True where depth is valid
├── extrinsics.npy # (100, 4, 4) float32 — world→camera (w2c)
├── intrinsics.npy # (100, 3, 3) float32 — pinhole K
└── meta.json # scene_id, frame_ids (parallel to array index), image_size
The first axis of every array is the frame, in the same order as meta.json's
frame_ids and the sorted images/ files.
Conventions
- Coordinate frame: OpenCV (x-right, y-down, z-forward).
extrinsicsis the world→camera (w2c) matrix; invert it for camera→world. - Depth: projective z-depth in metres (distance along the camera z-axis,
not Euclidean ray length). Use
valid_maskto ignore missing values. - Intrinsics: fixed per scene —
fx=fy=600,cx=599.5,cy=339.5.
Quick start
import tarfile, json, numpy as np
from huggingface_hub import hf_hub_download
p = hf_hub_download("helioom/3dvlm-replica_subset", "replica/office0.tar", repo_type="dataset")
tarfile.open(p).extractall("replica/")
meta = json.load(open("replica/office0/meta.json"))
depth = np.load("replica/office0/depth.npy") # (100, 680, 1200)
K = np.load("replica/office0/intrinsics.npy") # (100, 3, 3)
w2c = np.load("replica/office0/extrinsics.npy") # (100, 4, 4)
License
Replica is released for research use only under its original license; the same terms apply to this derived subset. See the Replica dataset.
- Downloads last month
- -