Image-To-3D
hypernet-image-to-3d-data
Hypernet Image-to-3D Dataset
Companion dataset to the hypernet and image-to-3d-deepsdf repositories.
Contents
Path
Size
Description
manifest.json
<1 MB
obj_idx ↔ uid ↔ LVIS category mapping for 1000 shapes
views.json
<1 KB
64 Fibonacci-sphere camera poses
watertight/
~21 GB
985 watertight .obj meshes (filename = uid hash)
sdf_samples/
~3 GB
976 .npz files, each with 200K (point, sdf) pairs
multiview/
~1.3 GB
976 .tar files; each tarball contains 64 PNG… See the full description on the dataset page: https://huggingface.co/datasets/bobthebuilderinternational/hypernet-image-to-3d-data.image-to-3D-eval-stability-under-gravityimage-to-3d185-hard-image-to-3D-eval-dataset
🏆 185-Hard: A Scenario-Driven Benchmark for VLM-Supervised Image-to-3D Generation
📖 Dataset Summary
185-Hard is a scenario-driven benchmark designed for industrial-grade Image-to-3D generation.
To address the gap between traditional metrics and human
judgment in Image-to-3D generation, we introduce a
VLM-based pipeline functioning as ’Senior 3D Art
Supervisors’ to assess geometry, texture, and adherence.
We present 185-hard, a hierarchical dataset covering… See the full description on the dataset page: https://huggingface.co/datasets/molodata/185-hard-image-to-3D-eval-dataset.