Pith Number

pith:2S42JA4V

pith:2026:2S42JA4VLDR5PAITG3DGWPYD6F

not attested not anchored not stored refs pending

InHabit: Leveraging Image Foundation Models for Scalable 3D Human Placement

Anna Khoreva, Gerard Pons-Moll, Istv\'an S\'ar\'andi, Jiayi Wang, Nikita Kister, Pradyumna YM

InHabit automatically generates large-scale 3D data of humans interacting with scenes by chaining 2D vision models to propose actions, insert figures, and optimize the results into scene-aligned SMPL-X bodies.

arxiv:2604.19673 v2 · 2026-04-21 · cs.CV

Open paper page JSON Open Graph Bundle Merged state Verified badge What is a Pith Number?

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{2S42JA4VLDR5PAITG3DGWPYD6F}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp

2 Internet Archive

3 Author claim open · sign in to claim

4 Citations open

5 Replications open

✓ Portable graph bundle live · download bundle · merged state

The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

Applied to Habitat-Matterport3D, InHabit produces the first large-scale photorealistic 3D human-scene interaction dataset, containing 78K samples across 800 building-scale scenes with complete 3D geometry, SMPL-X bodies, and RGB images. Augmenting standard training data with our samples improves RGB-based 3D human-scene reconstruction and contact estimation, and in a perceptual user study our data is preferred in 78% of cases over the state of the art.

C2weakest assumption

The assumption that off-the-shelf vision-language models will propose contextually meaningful actions and image-editing models will insert humans such that the subsequent optimization procedure can reliably produce physically plausible SMPL-X bodies aligned with scene geometry without artifacts or implausible configurations.

C3one line summary

InHabit generates 78K photorealistic 3D human-scene interaction samples across 800 scenes by rendering scenes, using foundation models to propose actions and insert humans, then optimizing to SMPL-X bodies, improving 3D reconstruction and contact estimation.

Receipt and verification

First computed	2026-05-27T02:06:14.109816Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

d4b9a4839558e3d7811336c66b3f03f16bbb8315ca6b3c66d4ccb160041c791f

Aliases

arxiv: 2604.19673 · arxiv_version: 2604.19673v2 · doi: 10.48550/arxiv.2604.19673 · pith_short_12: 2S42JA4VLDR5 · pith_short_16: 2S42JA4VLDR5PAIT · pith_short_8: 2S42JA4V

Agent API

Resolver JSON Graph JSON Events JSON Schema Signing key

Verify this Pith Number yourself

curl -sH 'Accept: application/ld+json' https://pith.science/pith/2S42JA4VLDR5PAITG3DGWPYD6F \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: d4b9a4839558e3d7811336c66b3f03f16bbb8315ca6b3c66d4ccb160041c791f

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "19c6738745ced72cc9a5974969d6e5051ce8a23f3943723292197f7aaf1ae289",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CV",
    "submitted_at": "2026-04-21T16:53:18Z",
    "title_canon_sha256": "fc4ef2de54a668dd4f30e93bfcff50fe7904f3fb889485050edebf050ad5189b"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2604.19673",
    "kind": "arxiv",
    "version": 2
  }
}