DTC-Synthetic — per-case data packages

DTC-Synthetic per-case data packages for the 7 scenes used in the paper (one zip per case).

Each DTC-Synthetic_<scene>.zip extracts to <scene>/ in our format, ready for the pipeline in README_DATA.md:

• cameras.json, synthetic_exrs_800/, mask_800/ — multi-view inputs
• mesh_transformed.obj + material.mtl, env.exr, diffrender_mode.json
• relit/ — ground-truth relit images (envmap12/3/6, 20 views) for the relight metric

Extract under data/dtc/adt_dslr/, then run the G-buffer prediction + Stage 2/3 as documented.