JiHyuk-Byun/3D-PAQA
3D-PAQA — Preference-Aligned 3D Quality Assessment Preference-aligned perceptual quality labels for 240,636 Objaverse assets, rated on six perceptual criteria. The goal of 3D-PAQA is to move beyond synthetic-distortion 3D-QA benchmarks and provide human-preference-aligned quality scores for real, human-created 3D assets, at a scale usable for training and benchmarking automatic quality evaluators. Drawn from a 264,966-asset Objaverse corpus. train.csv — 216,540 labeled assets… See the full description on the dataset page: https://huggingface.co/datasets/JiHyuk-Byun/3D-PAQA.
docs: link training code repo
Use generic @misc citation (no thesis reference)
Fix card: correct labeling method (MLLM exemplar-anchored RR, not raw human MOS) and scope (preference-aligned 3D-QA of Objaverse assets; generation-eval is unvalidated)
Document normalization: labels raw on purpose + criteria_stats.json
Add per-criterion label stats for optional downstream normalization
Fix acronym: Human Preference-Aligned Quality Assessment
Initial release: 3D-PAQA labels + PTv3 evaluator
initial commit