pith:HL2AJVZV
StereoPolicy: Improving Robotic Manipulation Policies via Stereo Perception
StereoPolicy processes synchronized stereo image pairs with 2D encoders and a fusion transformer to improve robotic manipulation policies without explicit 3D reconstruction.
arxiv:2605.09989 v2 · 2026-05-11 · cs.RO · cs.CV
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{HL2AJVZVG6OORA63STT5TZ5UEN}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
StereoPolicy delivers consistent improvements over RGB, RGB-D, point cloud, and multi-view baselines across three simulation benchmarks (RoboMimic, RoboCasa, OmniGibson) and is validated on real-robot tabletop and bimanual mobile manipulation tasks.
That processing each stereo image independently with pretrained 2D encoders and fusing via a Stereo Transformer is sufficient to implicitly capture spatial correspondence and disparity cues without explicit 3D reconstruction or camera calibration.
StereoPolicy fuses stereo image pairs via a Stereo Transformer on pretrained 2D encoders to boost robotic manipulation policies, showing gains over monocular, RGB-D, point cloud, and multi-view methods in simulations and real-robot tests.
Receipt and verification
| First computed | 2026-06-05T00:13:46.928048Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
3af404d735379ce883db94e7d9e7b42377d1d1ac629ee2781ed8937fc82437ce
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/HL2AJVZVG6OORA63STT5TZ5UEN \
| 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: 3af404d735379ce883db94e7d9e7b42377d1d1ac629ee2781ed8937fc82437ce
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "c60fb6f557a94350ec4b26b7ef8e9910b19e778df40055bdb126efb04363960a",
"cross_cats_sorted": [
"cs.CV"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"primary_cat": "cs.RO",
"submitted_at": "2026-05-11T05:06:12Z",
"title_canon_sha256": "67f9e5b3eb95d7e01569922077eb53cfa87efbc6b9f444eec2efd75d0abc716a"
},
"schema_version": "1.0",
"source": {
"id": "2605.09989",
"kind": "arxiv",
"version": 2
}
}