SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2302.14268 · ICLR · 2023

Self-Supervised Category-Level Articulated Object Pose Estimation with Part-Level SE(3) Equivariance

Yi Li, Ji Zhang, Xueyi Liu, Ruizhen Hu, Haibin Huang, He Wang

arXiv · PDF · Open in the Atlas

Code that ran

12 of 13 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 11 have no contract verdict; 0 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2302.14268.json · badge: /api/ran/2302.14268.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
Meowuu7/equi-articulated-pose canonical (classifier label)
classifier basis not recorded on the link
12 ran · 1 no recorded run
FunctionStatusWhere it lives
acos_safe Ran
ran on a synthesized input
Meowuu7/equi-articulated-pose/SPConvNets/pose_utils.py
code served (permissive licence) · get_code("a87a955c3ddcd7ae")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# Create a small tensor with values in a range suitable for acos
# acos expects input in [-1, 1], so we'll create values in that range
x = torch.tensor([
    [0.5, -0.3, 0.9, -0.1],
    [0.0, 0.8, -0.7, 0.2]
], dtype=torch.float32)

eps = 0.0001

ARGS = {
    "x": x,
    "eps": eps
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
angle_from_R Ran
ran on a synthesized input
Meowuu7/equi-articulated-pose/SPConvNets/pose_utils.py
code served (permissive licence) · get_code("ec94ea67605aa7f0")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 1, dtype family float, shape (2,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# The function angle_from_R takes a rotation matrix tensor R
# It computes einsum over batch and diagonal indices, so R should be shape (batch, dim, dim)
# A rotation matrix is typically square, so we'll use (2, 4, 4) for batch=2, dim=4

# Create a valid rotation-like matrix (doesn't need to be perfectly orthogonal for this test)
# Shape: (batch_size=2, dim=4, dim=4)
R = torch.randn(2, 4, 4)

# Normalize to make it closer to a valid rotation matrix
# (though the function doesn't strictly require it to be valid)
U, _, Vt = torch.svd(R)
R = torch.bmm(U, Vt)

ARGS = {
    "R": R
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 1, dtype family float, shape (2,) — shape (2,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
compute_rotation_matrix_from_angle Ran
ran on a synthesized input
Meowuu7/equi-articulated-pose/SPConvNets/model_utils.py
code served (permissive licence) · get_code("2d667b041476d287")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 4, dtype family float, shape (4, 2, 3, 3) as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import math

torch.manual_seed(0)

# Create anchor rotation matrices: na x 3 x 3
# We'll create 2 anchor matrices (na=2)
anchors = torch.eye(3).unsqueeze(0).repeat(2, 1, 1).float()
# Slightly perturb the second one to make it different
anchors[1] = torch.tensor([
    [0.9, -0.436, 0.0],
    [0.436, 0.9, 0.0],
    [0.0, 0.0, 1.0]
], dtype=torch.float32)

# Create angles: batch_size x na x 1
# batch_size = 4, na = 2, so shape is 4 x 2 x 1
angles = torch.randn(4, 2, 1).float()

# defined_axis is optional, set to None to let the function compute it
defined_axis = None

ARGS = {
    "anchors": anchors,
    "angles": angles,
    "defined_axis": defined_axis,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (4, 2, 3, 3) — shape (4, 2, 3, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
decode_rotation_info Ran
ran on a synthesized input
Meowuu7/equi-articulated-pose/SPConvNets/MotionDataset.py
code served (permissive licence) · get_code("1405a042ee5ad252")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 2, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

ARGS = {
    "rotate_info_encoding": 1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 2, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
from_rotation_mtx_to_axis Ran
ran on a synthesized input
Meowuu7/equi-articulated-pose/SPConvNets/model_utils.py
code served (permissive licence) · get_code("0cf06c572f8da0c0")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (2, 3) as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import math

torch.manual_seed(0)

# Create a batch of rotation matrices
# Shape: (2, 3, 3) - batch of 2 rotation matrices
# We'll create valid rotation matrices by using orthogonal matrices
batch_size = 2
rots = torch.eye(3).unsqueeze(0).repeat(batch_size, 1, 1)

# Add a small valid rotation matrix for diversity
angle = math.pi / 6  # 30 degrees
cos_a = math.cos(angle)
sin_a = math.sin(angle)
rot2 = torch.tensor([
    [cos_a, -sin_a, 0],
    [sin_a, cos_a, 0],
    [0, 0, 1]
], dtype=torch.float32)

rots[1] = rot2

ARGS = {
    "rots": rots
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 3) — shape (2, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
mean_angular_error Ran
ran on a synthesized input
Meowuu7/equi-articulated-pose/SPConvNets/pose_utils.py
code served (permissive licence) · get_code("8e0107376c1c1a26")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 1, dtype family float, shape (2,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# Create batch of rotation matrices
# R should be shape (batch_size, 3, 3)
batch_size = 2
pred_R = torch.randn(batch_size, 3, 3)
gt_R = torch.randn(batch_size, 3, 3)

# Ensure they are valid rotation matrices by orthogonalization
# Using QR decomposition to get orthogonal matrices
pred_R_ortho, _ = torch.linalg.qr(pred_R)
gt_R_ortho, _ = torch.linalg.qr(gt_R)

ARGS = {
    'pred_R': pred_R_ortho,
    'gt_R': gt_R_ortho,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 1, dtype family float, shape (2,) — shape (2,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
objective_eval Ran
ran on a synthesized input
Meowuu7/equi-articulated-pose/SPConvNets/eval_utils.py
code served (permissive licence) · get_code("35563dab8d13d779")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 1, dtype family float, shape (18,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Set random seed for reproducibility
np.random.seed(0)

# Create small test arrays for the function parameters
# params: 6 elements (3 for rotvec0, 3 for rotvec1)
params = np.array([0.1, 0.2, 0.3, 0.15, 0.25, 0.35], dtype=np.float64)

# x0: points to rotate (batch of 2, 3D points)
x0 = np.array([[1.0, 0.0, 0.0],
               [0.0, 1.0, 0.0]], dtype=np.float64)

# y0: target points after rotation (same shape as x0)
y0 = np.array([[0.95, 0.1, 0.05],
               [0.05, 0.95, 0.1]], dtype=np.float64)

# x1: another set of points (batch of 2, 3D points)
x1 = np.array([[1.0, 1.0, 0.0],
               [0.0, 1.0, 1.0]], dtype=np.float64)

# y1: target points after rotation (same shape as x1)
y1 = np.array([[0.9, 1.05, 0.1],
               [0.1, 0.9, 1.05]], dtype=np.float64)

# joints: shared joint points (batch of 2, 3D points)
joints = np.array([[0.5, 0.5, 0.5],
                   [0.2, 0.3, 0.4]], dtype=np.float64)

# isweight: boolean flag
isweight = True

ARGS = {
    'params': params,
    'x0': x0,
    'y0': y0,
    'x1': x1,
    'y1': y1,
    'joints': joints,
    'isweight': isweight
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (18,) — shape (18,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
pc_normalize Ran
ran, honoured the contract we drafted
Meowuu7/equi-articulated-pose/SPConvNets/MotionDataset.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from aboulch/ConvPoint, is pointer only (licence: NOASSERTION) · get_code("ec413739d406e611")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from aboulch/ConvPoint, and GitHub could not classify the licence of that repository (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

random_choice_noreplace Ran
ran on a synthesized input
Meowuu7/equi-articulated-pose/SPConvNets/ransac.py
code served (permissive licence) · get_code("2a51c1d89a4dc627")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 2, dtype family int, shape (5, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Set random seed for reproducibility
np.random.seed(0)

# Construct arguments for random_choice_noreplace
l = list(range(10))  # 1-D list to choose from
n_sample = 3  # sample size for each draw
num_draw = 5  # number of draws

ARGS = {
    "l": l,
    "n_sample": n_sample,
    "num_draw": num_draw,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family int, shape (5, 3) — shape (5, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
ransac_fit_t Ran
ran on a synthesized input
Meowuu7/equi-articulated-pose/SPConvNets/ransac.py
code served (permissive licence) · get_code("70119aeba1c66091")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# Create small batch tensors for batch_dt and batch_dr
# batch_dt: shape (batch_size, embedding_dim) - translation vectors
# batch_dr: shape (batch_size, embedding_dim) - rotation vectors (not used in function but required param)
batch_dt = torch.randn(2, 4)
batch_dr = torch.randn(2, 3)

# delta_r: not used in the function body, but required parameter
delta_r = 0.1

# max_iter: number of RANSAC iterations
max_iter = 10

# thres: threshold for inlier classification
thres = 0.5

ARGS = {
    "batch_dt": batch_dt,
    "batch_dr": batch_dr,
    "delta_r": delta_r,
    "max_iter": max_iter,
    "thres": thres
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
rotate_points_with_rotvec Ran
ran on a synthesized input
Meowuu7/equi-articulated-pose/SPConvNets/eval_utils.py
code served (permissive licence) · get_code("3451ac9a517b2831")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 2, dtype family float, shape (2, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create small sample data for the function
# points: array of 3D points to rotate, shape (batch_size, 3)
# rot_vecs: rotation vectors, shape (batch_size, 3)

batch_size = 2
num_dims = 3

# Create sample points
points = np.random.randn(batch_size, num_dims).astype(np.float64)

# Create sample rotation vectors
rot_vecs = np.random.randn(batch_size, num_dims).astype(np.float64)

ARGS = {
    "points": points,
    "rot_vecs": rot_vecs,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (2, 3) — shape (2, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
rotate_pts Ran
ran on a synthesized input
Meowuu7/equi-articulated-pose/SPConvNets/eval_utils.py
code served (permissive licence) · get_code("0dc8184ded8d6afa")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 2, dtype family float, shape (3, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create two point clouds with N=4 points in 3D space
np.random.seed(0)
source = np.random.randn(4, 3).astype(np.float32)
target = np.random.randn(4, 3).astype(np.float32)

ARGS = {
    "source": source,
    "target": target,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (3, 3) — shape (3, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
padding_1 Not yet run
unverified (harvested, no recorded run)
Meowuu7/equi-articulated-pose/SPConvNets/MotionDataset.py
code served (permissive licence) · get_code("76444ba99363e9ef")
How it ran

no invocation recorded.

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 1 later) and 12 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
pc_normalize 36 other papers
17 dated earlier, 19 later
1803.05827 (2018-03)
Local Spectral Graph Convolution for Point Set Feature Learning
1904.02375 (2019-04)
ConvPoint: Continuous Convolutions for Point Cloud Processing
1904.10014 (2019-04)
Linked Dynamic Graph CNN: Learning on Point Cloud via Linking Hierarchical Features
and 33 more not named here

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Category-level articulated object pose estimation aims to estimate a hierarchy of articulation-aware object poses of an unseen articulated object from a known category. To reduce the heavy annotations needed for supervised learning methods, we present a novel self-supervised strategy that solves this problem without any human labels. Our key idea is to factorize canonical shapes and articulated object poses from input articulated shapes through part-level equivariant shape analysis. Specifically, we first introduce the concept of part-level SE(3) equivariance and devise a network to learn features of such property. Then, through a carefully designed fine-grained pose-shape disentanglement strategy, we expect that canonical spaces to support pose estimation could be induced automatically. Thus, we could further predict articulated object poses as per-part rigid transformations describing how parts transform from their canonical part spaces to the camera space. Extensive experiments demonstrate the effectiveness of our method on both complete and partial point clouds from synthetic and real articulated object datasets. The project page with code and more information can be found at: equi-articulated-pose.github.io.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2302.14268")
get_code_for_paper("2302.14268")
have("2302.14268")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2302.14268.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2302.14268.svg)](https://syntology.ai/paper/2302.14268)

Connect an agent — have() is free.