Dataset Viewer
The dataset could not be loaded because the splits use different data file formats, which is not supported. Read more about the splits configuration. Click for more details.
Couldn't infer the same data file format for all splits. Got {NamedSplit('train'): ('webdataset', {}), NamedSplit('validation'): ('json', {}), NamedSplit('test'): ('json', {})}
Error code: FileFormatMismatchBetweenSplitsError
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.
dinov3-depth-packed
Packed WebDataset shards (shard-*.tar + manifest.json) for three depth-estimation datasets,
used to train/eval a simplified DINOv3 reimplementation. Original resolution, no resizing --
packing only repackages each dataset's existing raw images/depth maps into tar shards, it does
not alter pixel resolution.
Layout
diode/train/shard-*.tar, diode/train/manifest.json
diode/val/shard-*.tar, diode/val/manifest.json
hypersim/all/shard-*.tar, hypersim/all/manifest.json
kitti/train/shard-*.tar, kitti/train/manifest.json
kitti/val/shard-*.tar, kitti/val/manifest.json
kitti/test/shard-*.tar, kitti/test/manifest.json
Provenance
Each dataset was downloaded from its original public source and packed with a small per-dataset script (not included in this repo -- see the source project):
- DIODE:
download_diode.sh+pack_diode.py - Hypersim:
download_hypersim.py+pack_hypersim.py - KITTI:
pack_kitti.py(Eigen split, raw KITTI +data_depth_annotatedGT)
Raw (unpacked) source files are not included here -- only the packed WebDataset shards actually used for training.
- Downloads last month
- 18