SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2003.14111 · 2020

Disentangling and Unifying Graph Convolutions for Skeleton-Based Action Recognition

arXiv · PDF · Open in the Atlas

Code that ran

3 of 3 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; 2 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 1 failure of our instrument; 0 no recorded run), measured 2026-11-21; 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/2003.14111.json · badge: /api/ran/2003.14111.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
kenziyuliu/ms-g3d canonical (classifier label)
classifier basis: classifier_2025
3 ran · 0 no recorded run
FunctionStatusWhere it lives
get_nonzero_std Ran
ran, honoured the contract we drafted
kenziyuliu/ms-g3d/data_gen/ntu_gendata.py
code served (permissive licence) · get_code("47a234b123ed6809")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned Tensor, ndim 0, dtype family float, shape () 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

torch.manual_seed(0)

# The function expects s to be a 3D tensor with shape (batch, dim1, 3)
# where the last dimension has size 3 (for indexing [:, :, 0], [:, :, 1], [:, :, 2])
# 
# s.sum(-1).sum(-1) produces a 1D tensor of shape (batch,)
# index filters the batch dimension
# After filtering, it accesses s[:, :, 0], s[:, :, 1], s[:, :, 2]

# Create a tensor with shape (batch=2, dim1=4, dim2=3)
s = torch.randn(2, 4, 3)

ARGS = {'s': s}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
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 0, dtype family float, shape () — shape () 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
read_skeleton_filter Ran
ran; our contract draft was wrong, not the code
kenziyuliu/ms-g3d/data_gen/ntu_gendata.py
code served (permissive licence) · get_code("c6381bb48f81a144")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned dict. 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 tempfile
import os

# Create a temporary file with valid skeleton data
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt')

# Write sample skeleton data
# numFrame
temp_file.write("2\n")

# Frame 0
temp_file.write("1\n")  # numBody
# Body 0 info: bodyID clipedEdges handLeftConfidence handLeftState handRightConfidence handRightState isResticted leanX leanY trackingState
temp_file.write("1 0 1.0 2 0.9 3 0 0.1 0.2 2\n")
temp_file.write("3\n")  # numJoint
# Joint 0: x y z depthX depthY colorX colorY orientationW orientationX orientationY orientationZ trackingState
temp_file.write("0.1 0.2 0.3 100 200 300 400 1.0 0.0 0.0 0.0 2\n")
# Joint 1
temp_file.write("0.4 0.5 0.6 110 210 310 410 0.9 0.1 0.0 0.0 2\n")
# Joint 2
temp_file.write("0.7 0.8 0.9 120 220 320 420 0.8 0.2 0.1 0.0 2\n")

# Frame 1
temp_file.write("2\n")  # numBody
# Body 0
temp_file.write("2 0 0.8 1 0.7 2 0 0.0 0.1 2\n")
temp_file.write("2\n")  # numJoint
temp_file.write("1.1 1.2 1.3 130 230 330 430 1.0 0.0 0.0 0.0 2\n")
temp_file.write("1.4 1.5 1.6 140 240 340 440 0.95 0.05 0.0 0.0 2\n")
# Body 1
temp_file.write("3 1 0.6 0 0.5 1 0 0.2 0.3 2\n")
temp_file.write("2\n")  # numJoint
temp_file.write("2.1 2.2 2.3 150 250 350 450 0.9 0.1 0.1 0.1 2\n")
temp_file.write("2.4 2.5 2.6 160 260 360 460 0.85 0.15 0.05 0.05 2\n")

temp_file.close()

ARGS = {
    'file': temp_file.name
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not 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
read_xyz Ran
ran, honoured the contract we drafted
kenziyuliu/ms-g3d/data_gen/ntu_gendata.py
code served (permissive licence) · get_code("7e6ab20b3ba9e367")
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) and returned ndarray, ndim 4, dtype family float, shape (3, 2, 25, 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 numpy as np
import tempfile
import os

# Create a temporary skeleton file with valid format
def create_temp_skeleton_file():
    temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.skeleton')
    
    # Write number of frames
    temp_file.write("2\n")  # 2 frames
    
    # Frame 1
    temp_file.write("1\n")  # 1 body
    # Body info (10 values)
    temp_file.write("0 0 0.5 0 0.5 0 0 0.1 0.2 2\n")
    temp_file.write("25\n")  # 25 joints
    # Write 25 joints (12 values each)
    for j in range(25):
        joint_line = " ".join([str(float(i)) for i in range(12)])
        temp_file.write(joint_line + "\n")
    
    # Frame 2
    temp_file.write("1\n")  # 1 body
    # Body info (10 values)
    temp_file.write("0 0 0.5 0 0.5 0 0 0.1 0.2 2\n")
    temp_file.write("25\n")  # 25 joints
    # Write 25 joints (12 values each)
    for j in range(25):
        joint_line = " ".join([str(float(i)) for i in range(12)])
        temp_file.write(joint_line + "\n")
    
    temp_file.close()
    return temp_file.name

skeleton_file = create_temp_skeleton_file()

ARGS = {
    'file': skeleton_file,
    'max_body': 4,
    'num_joint': 25
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 4, dtype family float, shape (3, 2, 25, 2) — shape (3, 2, 25, 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

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 3 later) and 0 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
get_nonzero_std 4 other papers
1 dated earlier, 3 later
1904.12659 (2019-04)
Actional-Structural Graph Convolutional Networks for Skeleton-based Action Recognition
2104.00946 (2021-04)
UAV-Human: A Large Benchmark for Human Behavior Understanding with Unmanned Aerial Vehicles
2108.04536 (2021-08)
Learning Multi-Granular Spatio-Temporal Graph Network for Skeleton-based Action Recognition
and 1 more not named here
read_skeleton_filter 4 other papers
1 dated earlier, 3 later
1904.12659 (2019-04)
Actional-Structural Graph Convolutional Networks for Skeleton-based Action Recognition
2104.00946 (2021-04)
UAV-Human: A Large Benchmark for Human Behavior Understanding with Unmanned Aerial Vehicles
2108.04536 (2021-08)
Learning Multi-Granular Spatio-Temporal Graph Network for Skeleton-based Action Recognition
and 1 more not named here
read_xyz 3 other papers
1 dated earlier, 2 later
1904.12659 (2019-04)
Actional-Structural Graph Convolutional Networks for Skeleton-based Action Recognition
2108.04536 (2021-08)
Learning Multi-Granular Spatio-Temporal Graph Network for Skeleton-based Action Recognition
aaai_25127 (2023 (year only))
Hierarchical Contrast for Unsupervised Skeleton-Based Action Representation Learning

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 2026-11-21.

Repositories linked to this paper

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

Abstract

Spatial-temporal graphs have been widely used by skeleton-based action recognition algorithms to model human action dynamics. To capture robust movement patterns from these graphs, long-range and multi-scale context aggregation and spatial-temporal dependency modeling are critical aspects of a powerful feature extractor. However, existing methods have limitations in achieving (1) unbiased long-range joint relationship modeling under multi-scale operators and (2) unobstructed cross-spacetime information flow for capturing complex spatial-temporal dependencies. In this work, we present (1) a simple method to disentangle multi-scale graph convolutions and (2) a unified spatial-temporal graph convolutional operator named G3D. The proposed multi-scale aggregation scheme disentangles the importance of nodes in different neighborhoods for effective long-range modeling. The proposed G3D module leverages dense cross-spacetime edges as skip connections for direct information propagation across the spatial-temporal graph. By coupling these proposals, we develop a powerful feature extractor named MS-G3D based on which our model outperforms previous state-of-the-art methods on three large-scale datasets: NTU RGB+D 60, NTU RGB+D 120, and Kinetics Skeleton 400.

For agents

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

get_harvested_code_for_paper("2003.14111")
get_code_for_paper("2003.14111")
have("2003.14111")

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

curl https://syntology.ai/api/ran/2003.14111.json

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

[![Syntology run record](https://syntology.ai/api/ran/2003.14111.svg)](https://syntology.ai/paper/2003.14111)

Connect an agent — have() is free.