遇见数据集

ODD INN multistate photogrammetry and Unreal Engine evidence and retrospective analyses

收藏
Zenodo2026-09-24 更新2026-10-01 收录
官方服务:

资源简介:

Selected original technical records from the completed ODD INN research project, exported validation notebooks, de-identified quality-assurance observations and retrospective arithmetic analyses supporting two articles on multistate photogrammetry and Unreal Engine 5 state and material transitions. Version 1.0.1 is a correction of version 1.0.0. It fixes optimization-sensitive input and integrity checks, makes UTF-8 encoding explicit, normalizes NumPy integral parameters, and adds targeted synthetic regression checks. It also adds a critical note identifying a conversational residue, an incorrect DOI assignment and unsupported general colour-perception thresholds in historical source P11. Original source bytes and historical observations are preserved. These changes are not new project measurements or corrected historical image-accuracy results. See CHANGELOG.md and P11_source_critical_note.md. DOWNLOAD INSTRUCTIONS: Start with START_HERE.md. The evidence archive is supplied as binary parts. Download every numbered part, PARTS_MANIFEST.json and reassemble.py into one folder, then run python3 reassemble.py. The helper verifies every part and the complete archive before completing the ZIP. Original documentary contributors are credited in retained reports. The package does not contain a complete final Blender addon, an editable Unreal project, original full-resolution validation images or raw frame-time series. It supports documentary review and arithmetic reanalysis, not full experimental replication. Technical project records are not peer-reviewed publications; P13–P15 are provided only as selected technical extracts. Prospective protocols contain no executed trials. Underlying project funding: NCBR, European Regional Development Fund, Smart Growth Operational Programme 2014–2020, grant POIR.01.01.01-00-0280/22-00. Preparation of the articles received no separate funding. No funder influenced the present analysis or the decision to publish the articles. Approved component licenses: CC BY 4.0 for documentation and data, MIT for the explicitly listed new analysis, integrity-checking and reassembly scripts. LICENSE_SCOPE.md distinguishes historical code and third-party material. The depositor approved these terms and publication on 24 September 2026. SHA256SUMS.txt and FILE_MANIFEST.json support integrity checking. REPRODUCING.md describes analyses rerun; PRIVACY_REVIEW.md and privacy_redactions.json document narrowly redacted evidence copies. Original source and deposited-file hashes remain distinct.

提供机构:
Zenodo
创建时间:
2026-09-24
二维码
社区交流群
二维码
科研交流群
商业服务