HiRoom.

HiRoom — a Blender-rendered synthetic indoor dataset of 30 furnished/living-room scenes used by the authors of "Depth Anything 3" (DA3) as part of their visual-geometry benchmark. The dataset is used for evaluating any-view geometry / 3D reconstruction and camera-pose estimation in the DA3 paper; the paper reports using a reconstruction F1 threshold d = 0.05 m for HiRoom. No standalone Hugging Face dataset page or separate publication for HiRoom was found; the dataset appears in the DA3 benchmark and tech report (arXiv:2511.10647 / project page).

Best published scores.

No results indexed yet — be the first to submit a score.

Neighbouring benchmarks.

Other Few-Shot Image Classification datasets on Codesota.

Have a score that beats
this table?

Submit a checkpoint and a reproduction script. We will run it, publish the score, and — if it takes the top — annotate the step on the progress chart with your name.

• 01A public checkpoint or API endpoint
• 02A reproduction script with frozen commit + seed
• 03Declared evaluation environment (Python, deps)
• 04One row per metric declared by this dataset
• 05A contact so we can follow up on discrepancies