| pretty_name: ZJU-MV | |
| tags: | |
| - multiview | |
| - dynamic-scene | |
| - 4d-reconstruction | |
| # ZJU-MV | |
| ZJU-MV is the dynamic multi-view companion dataset to ZJU-OLAT. | |
| The release is organized as one tar archive per synchronized sequence. Each | |
| archive contains per-view H.264 MP4 videos at 25 fps together with image/video | |
| size reports and per-view compression metadata. | |
| ## Repository layout | |
| ```text | |
| cases/ | |
| MV001_<case_name>/ | |
| MV001_<case_name>.tar | |
| MV001_<case_name>.tar.sha256 | |
| ``` | |
| Verify an archive after download: | |
| ```bash | |
| sha256sum -c MV001_<case_name>.tar.sha256 | |
| ``` | |
| Additional documentation, usage terms, and dataset statistics will be added as | |
| the release is finalized. | |