Here is the repo for a sample (11 scenes/videos) of the full DL3DV-10K dataset. Each scene has a raw 4K video and the processed colmap results.
5c3af581028068a3c402c7cbe16ecf9471ddf2897c34ab634b7b1b6cf81aba00
├── colmap
│ ├── colmap
│ │ ├── features.h5
│ │ ├── global-feats-netvlad.h5
│ │ ├── matches.h5
│ │ ├──… See the full description on the dataset page:
https://huggingface.co/datasets/DL3DV/DL3DV-10K-Sample.