Pith Number

pith:LHXJRE7H

pith:2026:LHXJRE7HKN6X56T5NJROFKTU34

not attested not anchored not stored refs resolved

Realiz3D: 3D Generation Made Photorealistic via Domain-Aware Learning

Andrea Vedaldi, Egor Zakharov, Ido Sobol, Kihyuk Sohn, Max Bluvstein, Or Litany, Yoav Blum

Realiz3D decouples visual domain from control signals via a co-variate and residual adapters so diffusion models can apply 3D controls without adopting synthetic appearance.

arxiv:2605.13852 v1 · 2026-03-25 · cs.GR · cs.CV · cs.LG

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{LHXJRE7HKN6X56T5NJROFKTU34}

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

We introduce Realiz3D, a lightweight framework for training diffusion models, that decouples controls and visual domain. The key idea is to explicitly learn visual domain, real or synthetic, separately from other control signals by introducing a co-variate that, fed into small residual adapters, shifts the domain.

C2weakest assumption

The domain gap largely arises from the model learning an unintended association between the presence of control signals and the synthetic appearance of the images, which the co-variate and adapters can fully mitigate without losing control accuracy.

C3one line summary

Realiz3D decouples visual domain from 3D controls in diffusion models via domain-aware residual adapters to enable photorealistic controllable generation.

References

46 extracted · 46 resolved · 5 Pith anchors

[1] Deep vit features as dense visual descriptors.arXiv preprint arXiv:2112.05814, 2(3):4 2021

[2] Meta 3D TextureGen: Fast and consistent texture generation for 3d objects 2024

[3] Synthlight: Por- trait relighting with diffusion model by learning to re-render synthetic faces 2025

[4] Still-moving: Customized video generation without customized video data.ACM Transactions on Graphics (TOG), 43(6):1–11, 2024 2024

[5] Ambient diffu- sion: Learning clean distributions from corrupted data 2023

Formal links

1 machine-checked theorem link

Receipt and verification

First computed	2026-05-17T23:39:19.593494Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

59ee9893e7537d7efa7d6a62e2aa74df3aaa0d013e25b93c0c56bacc25d089ec

Aliases

arxiv: 2605.13852 · arxiv_version: 2605.13852v1 · doi: 10.48550/arxiv.2605.13852 · pith_short_12: LHXJRE7HKN6X · pith_short_16: LHXJRE7HKN6X56T5 · pith_short_8: LHXJRE7H

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/LHXJRE7HKN6X56T5NJROFKTU34 \
  | 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: 59ee9893e7537d7efa7d6a62e2aa74df3aaa0d013e25b93c0c56bacc25d089ec

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "0445d383dfea12380fa8f9a75b165a33c1e31b321c48076d0016219e271aa124",
    "cross_cats_sorted": [
      "cs.CV",
      "cs.LG"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.GR",
    "submitted_at": "2026-03-25T11:30:22Z",
    "title_canon_sha256": "95f3416c3193a9313f442088e23f7bf49e841137d5cbc45da33098c61f62f69e"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.13852",
    "kind": "arxiv",
    "version": 1
  }
}