Dataset · Hugging Face

rabbitKabbit/steerscene-backbone-outputs

Name: steerscene-backbone-outputs
Creator: rabbitKabbit
License: other
Keywords: license:other, size_categories:n<1K, format:text, modality:3d, modality:text, library:datasets, library:mlcroissant, region:us, 3d-scene-synthesis, guidance, steerscene

SteerScene on other backbones: best outputs SteerScene training-free guidance applied to PhyScene (CVPR'24) and InstructScene (ICLR'24) without retraining. Box sizes are frozen during guidance (only positions and rotations are steered). Weights were tuned on held-out data (PhyScene seed 2028; InstructScene val) and the outputs here are on the reported sets. physcene/ (living room, seed 2027, 1024 scenes) steerscene_best/: collision_weight 100, room_weight 1.… See the full description on the dataset page: https://huggingface.co/datasets/rabbitKabbit/steerscene-backbone-outputs.

View on Hugging FaceReport an issue

Author / organization	rabbitKabbit
License (as declared)	other
Downloads	0 (as reported by Hugging Face)
Likes	0
Created	23 October 2026
Last updated at source	23 October 2026

3d-scene-synthesisguidancesteerscene

Embed this entry

<iframe src="https://silytics.si/embed/item/8d987e34-4ad3-400d-a554-bdbeb26d9c2d" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

More from rabbitKabbit

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 23 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.