SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2308.07313 · ICCV · 2023

Group Pose: A Simple Baseline for End-to-End Multi-person Pose Estimation

Jian Wang, Errui Ding, Yao Zhao, Jingdong Wang, Qiang Chen, Junyu Han, Kun Yao, Xiaolong Li, Huan Liu, Zichang Tan, Jiang-Jiang Liu, Xiangbo Su

arXiv · PDF · Open in the Atlas

Code that ran

21 of 22 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; 3 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 17 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/2308.07313.json · badge: /api/ran/2308.07313.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
Michel-liu/GroupPose-Paddle canonical (classifier label)
classifier basis: census
15 ran · 1 no recorded run
Michel-liu/GroupPose canonical (classifier label)
classifier basis: census
6 ran · 0 no recorded run
FunctionStatusWhere it lives
build_matcher Ran
ran; our contract draft was wrong, not the code
Michel-liu/GroupPose/models/grouppose/matcher.py
code served (permissive licence) · get_code("b470cc20ddf04258")
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 HungarianMatcher. 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
from torch import nn

torch.manual_seed(0)

class ArgsConfig:
    def __init__(self):
        self.matcher_type = 'HungarianMatcher'
        self.set_cost_class = 1.0
        self.focal_alpha = 0.25
        self.set_cost_keypoints = 1.0
        self.set_cost_oks = 0.01
        self.num_body_points = 17

ARGS = {
    'args': ArgsConfig()
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type HungarianMatcher — a HungarianMatcher output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a HungarianMatcher 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
dump_value Ran
ran; our contract draft was wrong, not the code
Michel-liu/GroupPose-Paddle/ppdet/core/workspace.py
code served (permissive licence) · get_code("891281774f8942bd")
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 str. 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 yaml

torch.manual_seed(0)

# The function takes a single parameter 'value' and processes it
# It checks if value has __dict__, or is a dict, tuple, or list
# Then it yaml.dump's it and does string processing
# So we need to test with various types of values

ARGS = {
    'value': {
        'key1': 'test_string',
        'key2': 42,
        'nested': {'inner': [1, 2, 3]},
        'tuple_field': (1, 2, 3),
        'list_field': [4, 5, 6]
    }
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, yaml — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
gen_encoder_output_proposals Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Michel-liu/GroupPose/models/grouppose/utils.py
code served (permissive licence) · get_code("e3b2768e6a9c1cb5")
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, ndim 4, 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 torch
from torch import nn, Tensor

torch.manual_seed(0)

# Define parameters for the function
# memory: bs, sum{hw}, d_model
# For 2 levels with spatial shapes (2, 3) and (2, 2): sum{hw} = 2*3 + 2*2 = 10
batch_size = 2
d_model = 4
total_hw = 10  # sum of all spatial dimensions

# Create memory tensor: (batch_size, total_hw, d_model)
memory = torch.randn(batch_size, total_hw, d_model)

# Create memory_padding_mask: (batch_size, total_hw)
# False means valid, True means padding
memory_padding_mask = torch.zeros(batch_size, total_hw, dtype=torch.bool)
# Mark some positions as padding
memory_padding_mask[0, 8:] = True
memory_padding_mask[1, 9:] = True

# Create spatial_shapes: (nlevel, 2)
# Two levels: first level is 2x3, second level is 2x2
spatial_shapes = torch.tensor([[2, 3], [2, 2]], dtype=torch.long)

# Create learnedwh: (2,) tensor
learnedwh = torch.randn(2)

ARGS = {
    "memory": memory,
    "memory_padding_mask": memory_padding_mask,
    "spatial_shapes": spatial_shapes,
    "learnedwh": learnedwh,
}
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 tuple, ndim 4, dtype family float — a tuple 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
gen_sineembed_for_position Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Michel-liu/GroupPose/models/grouppose/utils.py
code served (permissive licence) · get_code("4c82654ae6131ecc")
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 3, dtype family float, shape (2, 3, 256) 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 math

torch.manual_seed(0)

# The function expects pos_tensor with shape (batch, num_positions, 2 or 4)
# where the last dimension is either:
# - 2: (x, y) coordinates
# - 4: (y, x, w, h) coordinates

# Create a small pos_tensor with shape (2, 3, 2) for batch_size=2, num_positions=3, coords=2
pos_tensor_2d = torch.randn(2, 3, 2, dtype=torch.float32)

# Also test with 4D coordinates
pos_tensor_4d = torch.randn(2, 3, 4, dtype=torch.float32)

ARGS = {
    "pos_tensor": pos_tensor_2d
}
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 3, dtype family float, shape (2, 3, 256) — shape (2, 3, 256) 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
generate_scale Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Michel-liu/GroupPose-Paddle/deploy/auto_compression/paddle_inference_eval.py
code served (permissive licence) · get_code("c115c6190dd6b6d3")
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, ndim 1, 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

# Create a small test image with shape (height, width, channels)
im = np.random.randint(0, 256, size=(480, 640, 3), dtype=np.uint8)

# Define target shape (height, width)
target_shape = (224, 224)

# Create arguments dictionary
ARGS = {
    "im": im,
    "target_shape": target_shape,
    "keep_ratio": True
}
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 tuple, ndim 1, dtype family float — a tuple 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
get_affine_mat_kernel Ran
ran; our contract draft was wrong, not the code
Michel-liu/GroupPose-Paddle/ppdet/modeling/keypoint_utils.py
code served (permissive licence) · get_code("e9ea8d6169dd3c5a")
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 numpy as np
import cv2

# Define ARGS dictionary for get_affine_mat_kernel
ARGS = {
    'h': 256,
    'w': 192,
    's': 256,
    'inv': False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy — 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
get_affine_transform Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Michel-liu/GroupPose-Paddle/ppdet/modeling/keypoint_utils.py
code served (permissive licence) · get_code("2fc1e1b3d4511e08")
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 cv2
import numpy as np

ARGS = {
    'center': np.array([100.0, 100.0], dtype=np.float32),
    'input_size': np.array([200.0, 200.0], dtype=np.float32),
    'rot': 0.0,
    'output_size': np.array([64.0, 64.0], dtype=np.float32),
    'shift': (0.0, 0.0),
    'inv': False
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, 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
get_color_map_list Ran
ran, honoured the contract we drafted
Michel-liu/GroupPose-Paddle/deploy/auto_compression/paddle_inference_eval.py
code served (permissive licence) · get_code("6a8c58548a03d3a2")
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 int; 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 torch
import numpy as np

torch.manual_seed(0)

ARGS = {
    "num_classes": 5
}
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 list, ndim 2, dtype family int — 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
get_warp_matrix Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Michel-liu/GroupPose-Paddle/ppdet/modeling/keypoint_utils.py
code served (permissive licence) · get_code("75f2a6d60178f80d")
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

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

# Define arguments for get_warp_matrix
theta = 30.0  # Rotation angle in degrees

size_input = np.array([640, 480], dtype=np.float32)  # Input image size [w, h]

size_dst = np.array([256, 256], dtype=np.float32)  # Output image size [w, h]

size_target = np.array([200, 200], dtype=np.float32)  # ROI size in input plane [w, h]

ARGS = {
    "theta": theta,
    "size_input": size_input,
    "size_dst": size_dst,
    "size_target": size_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 (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
get_weights_url Ran
ran; our contract draft was wrong, not the code
Michel-liu/GroupPose-Paddle/ppdet/model_zoo/model_zoo.py
code served (permissive licence) · get_code("e241ea7c12e86306")
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 str. 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 os.path as osp

def get_weights_url(model_name):
    return 'ppdet://models/{}.pdparams'.format(osp.split(model_name)[-1])

# The function takes model_name and uses osp.split() on it to get the last component,
# then formats it into a URL string. model_name should be a string path.

ARGS = {
    'model_name': '/path/to/model_checkpoint'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
identity Ran
ran, honoured the contract we drafted
Michel-liu/GroupPose-Paddle/ppdet/modeling/ops.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from dholzmueller/universal_double_descent, is pointer only (licence: NONE) · get_code("9910e2fc297f8665")
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 dholzmueller/universal_double_descent, and GitHub detected no licence in that repository when it was checked. 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.

image_preprocess Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Michel-liu/GroupPose-Paddle/deploy/auto_compression/paddle_inference_eval.py
code served (permissive licence) · get_code("145a8b9a0a3f9d79")
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 cv2
import numpy as np
import tempfile
import os

# Create a temporary image file for testing
temp_dir = tempfile.mkdtemp()
img_path = os.path.join(temp_dir, "test_image.jpg")

# Create a simple test image (100x80 BGR image)
test_img = np.random.randint(0, 256, (100, 80, 3), dtype=np.uint8)
cv2.imwrite(img_path, test_img)

# Define arguments for image_preprocess function
ARGS = {
    "img_path": img_path,
    "target_shape": (224, 224)
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy — 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
load_config Ran
ran; our contract draft was wrong, not the code
Michel-liu/GroupPose-Paddle/ppdet/core/workspace.py
code served (permissive licence) · get_code("7b71ca54a440c129")
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 AttrDict. 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 os
import tempfile
import yaml

# Create a temporary directory and YAML files for testing
temp_dir = tempfile.mkdtemp()

# Create a base config file
base_config = {
    'model': 'base_model',
    'learning_rate': 0.001,
    'batch_size': 32
}

base_yml_path = os.path.join(temp_dir, 'base_config.yml')
with open(base_yml_path, 'w') as f:
    yaml.dump(base_config, f)

# Create a main config file that references the base
main_config = {
    '_BASE_': ['base_config.yml'],
    'model': 'main_model',
    'epochs': 10
}

main_yml_path = os.path.join(temp_dir, 'config.yml')
with open(main_yml_path, 'w') as f:
    yaml.dump(main_config, f)

# Need to import collections.abc for the dict_merge function
import collections.abc as collectionsAbc

# Define ARGS with the file_path parameter
ARGS = {
    'file_path': main_yml_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
yaml — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type AttrDict — a AttrDict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a AttrDict 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
multiclass_nms Ran
ran; our contract draft was wrong, not the code
Michel-liu/GroupPose-Paddle/ppdet/modeling/post_process.py
code served (permissive licence) · get_code("0cf81b30bc14579c")
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 3, 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

# Create sample bounding boxes with format: [class_id, score, x1, y1, x2, y2]
# We'll create 6 boxes across 2 classes
bboxs = np.array([
    [0, 0.9, 10, 10, 50, 50],    # class 0, high score
    [0, 0.8, 15, 15, 55, 55],    # class 0, overlapping with first
    [0, 0.7, 100, 100, 150, 150], # class 0, non-overlapping
    [1, 0.85, 20, 20, 60, 60],   # class 1, high score
    [1, 0.75, 25, 25, 65, 65],   # class 1, overlapping with previous
    [1, 0.6, 200, 200, 250, 250], # class 1, non-overlapping
], dtype=np.float32)

num_classes = 2
match_threshold = 0.6
match_metric = 'iou'

ARGS = {
    'bboxs': bboxs,
    'num_classes': num_classes,
    'match_threshold': match_threshold,
    'match_metric': match_metric,
}
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 3, 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
nms Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Michel-liu/GroupPose-Paddle/ppdet/modeling/post_process.py
code served (permissive licence) · get_code("c54757be76d21549")
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, 5) 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 sample detections with shape [N, 5] where columns are [score, x1, y1, x2, y2]
# Create a few bounding boxes with overlapping regions
dets = np.array([
    [0.9, 10.0, 20.0, 50.0, 60.0],  # High confidence box
    [0.8, 15.0, 25.0, 55.0, 65.0],  # Overlapping box (should be suppressed with high threshold)
    [0.7, 100.0, 100.0, 150.0, 150.0],  # Non-overlapping box
    [0.6, 102.0, 102.0, 152.0, 152.0],  # Slightly overlapping with previous
], dtype=np.float32)

ARGS = {
    'dets': dets,
    'match_threshold': 0.5,
    'match_metric': 'iou'
}
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, 5) — shape (2, 5) 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
norm_angle Ran
ran, violated the contract we drafted
Michel-liu/GroupPose-Paddle/ppdet/modeling/rbox_utils.py
code served (permissive licence) · get_code("6c7db6948ecffd97")
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 float, 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 numpy as np

# The function norm_angle takes two parameters:
# 1. angle - a numeric value (could be float, int, or array-like)
# 2. range - a list with two elements [min, max]

# Looking at the function body:
# - It performs arithmetic operations: (angle - range[0]) % range[1] + range[0]
# - range is indexed at [0] and [1]
# - The default range is [-np.pi / 4, np.pi]

# We'll test with a simple float angle value
angle_value = 0.5

# The range parameter has a default, but we can provide a custom one
range_value = [-np.pi / 4, np.pi]

ARGS = {
    "angle": angle_value,
    "range": range_value,
}
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 float, 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
poly2rbox_le135_np Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Michel-liu/GroupPose-Paddle/ppdet/modeling/rbox_utils.py
code served (permissive licence) · get_code("5b37cc101c76cbd0")
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 1, 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

# The function poly2rbox_le135_np takes a polygon (list/array of 8 coordinates)
# and converts it to a rotated bounding box representation.
# It expects: poly parameter which is array-like with at least 8 elements [x1, y1, x2, y2, x3, y3, x4, y4]

# Create a valid polygon - 4 points forming a quadrilateral
# Using a simple trapezoid-like shape
poly = [0.0, 0.0, 10.0, 2.0, 11.0, 12.0, 1.0, 10.0]

ARGS = {
    "poly": poly
}
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 1, 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
poly2rbox_oc_np Ran
ran, honoured the contract we drafted
Michel-liu/GroupPose-Paddle/ppdet/modeling/rbox_utils.py
code served (permissive licence) · get_code("155145950e1115ca")
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 1, 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
import cv2

# The function poly2rbox_oc_np takes a polygon as input (8 coordinates for 4 points)
# and converts it to a rotated bounding box representation

# Create a simple polygon: 4 points forming a quadrilateral
# Format: [x1, y1, x2, y2, x3, y3, x4, y4]
poly = [0.0, 0.0, 10.0, 0.0, 10.0, 5.0, 0.0, 5.0]

ARGS = {
    "poly": poly
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, 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
sigmoid_focal_loss Ran
ran; our contract draft was wrong, not the code
Michel-liu/GroupPose/models/grouppose/utils.py
code served (permissive licence) · get_code("5c0711aada67957e")
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 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
import torch.nn.functional as F

torch.manual_seed(0)

# Create small tensors for inputs and targets
# inputs: predictions (logits) - shape (2, 4) for batch_size=2, num_predictions=4
inputs = torch.randn(2, 4, requires_grad=True)

# targets: binary labels - shape (2, 4) matching inputs
targets = torch.randint(0, 2, (2, 4), dtype=torch.float32)

# num_boxes: number of boxes for normalization
num_boxes = 2

# alpha and gamma are already float parameters with defaults
alpha = 0.25
gamma = 2.0

ARGS = {
    "inputs": inputs,
    "targets": targets,
    "num_boxes": num_boxes,
    "alpha": alpha,
    "gamma": gamma,
}
Setup block drafted by
a language model, Claude Haiku 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
window_partition Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Michel-liu/GroupPose/models/grouppose/backbones/swin_transformer.py
code served (permissive licence) · get_code("f9fd6241d935f07b")
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 4, dtype family float, shape (12, 4, 4, 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

torch.manual_seed(0)

# Create a tensor with shape (B, H, W, C)
# B = 2 (batch size)
# H = 8 (height, must be divisible by window_size)
# W = 12 (width, must be divisible by window_size)
# C = 4 (channels)
# window_size = 4

B = 2
H = 8
W = 12
C = 4
window_size = 4

x = torch.randn(B, H, W, C)

ARGS = {
    'x': x,
    'window_size': window_size
}
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 4, dtype family float, shape (12, 4, 4, 4) — shape (12, 4, 4, 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
window_reverse Ran
ran; our contract draft was wrong, not the code
Michel-liu/GroupPose/models/grouppose/backbones/swin_transformer.py
code served (permissive licence) · get_code("fb32094c6dbece71")
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 4, dtype family float, shape (2, 8, 12, 6) 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)

# Define parameters
# We need windows of shape (num_windows*B, window_size, window_size, C)
# where num_windows = (H // window_size) * (W // window_size)

B = 2
H = 8
W = 12
window_size = 4
C = 6

# Calculate num_windows
num_windows_h = H // window_size  # 8 // 4 = 2
num_windows_w = W // window_size  # 12 // 4 = 3
num_windows = num_windows_h * num_windows_w  # 2 * 3 = 6

# Total number of windows across batch
total_windows = num_windows * B  # 6 * 2 = 12

# Create windows tensor
windows = torch.randn(total_windows, window_size, window_size, C)

ARGS = {
    'windows': windows,
    'window_size': window_size,
    'H': H,
    'W': W
}
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 4, dtype family float, shape (2, 8, 12, 6) — shape (2, 8, 12, 6) 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
dict_merge Not yet run
unverified (harvested, no recorded run)
Michel-liu/GroupPose-Paddle/ppdet/core/workspace.py
code served (permissive licence) · get_code("553169f24c2e1789")
How it ran

no invocation recorded.

Also present in other papers' code

Of 22 samples lifted from this paper's linked repositories, 22 carry a function identity and 0 do not. Of the 22 distinct functions with an identity, 18 are also present in other papers whose code we hold (13 with at least one paper dated earlier, 4 the same month, 15 later) and 4 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
window_reverse 227 other papers
105 dated earlier, 6 the same month, 107 later, 9 same year, order not determinable · 34 only as an assembly brick
1608.03983 (2016-08)
SGDR: Stochastic Gradient Descent with Warm Restarts
1711.10925 (2017-11)
Deep Image Prior
2003.13678 (2020-03)
Designing Network Design Spaces
and 224 more not named here
window_partition 223 other papers
100 dated earlier, 7 the same month, 107 later, 9 same year, order not determinable · 32 only as an assembly brick
1706.03762 (2017-06)
Attention Is All You Need
1711.10925 (2017-11)
Deep Image Prior
2003.13678 (2020-03)
Designing Network Design Spaces
and 220 more not named here
identity 69 other papers
40 dated earlier, 1 the same month, 27 later, 1 same year, order not determinable · 5 only as an assembly brick
1706.07296 (2017-06)
A Minimal Developmental Model Can Increase Evolvability in Soft Robots
1802.09419 (2018-02)
Stochastic Hyperparameter Optimization through Hypernetworks
1804.06055 (2018-04)
Co-occurrence Feature Learning from Skeleton Data for Action Recognition and Detection with Hierarchical Aggregation
and 66 more not named here
sigmoid_focal_loss 56 other papers
31 dated earlier, 2 the same month, 20 later, 3 same year, order not determinable · 1 only as an assembly brick
2005.12872 (2020-05)
End-to-End Object Detection with Transformers
2010.04159 (2020-10)
Deformable DETR: Deformable Transformers for End-to-End Object Detection
2103.11731 (2021-03)
Meta-DETR: Image-Level Few-Shot Object Detection with Inter-Class Correlation Exploitation
and 53 more not named here
gen_sineembed_for_position 7 other papers
4 dated earlier, 3 later
2203.01305 (2022-03)
DN-DETR: Accelerate DETR Training by Introducing Query DeNoising
2203.03605 (2022-03)
DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection
2206.02777 (2022-06)
Mask DINO: Towards A Unified Transformer-based Framework for Object Detection and Segmentation
and 4 more not named here
dump_value 4 other papers
3 dated earlier, 1 later
1912.06218 (2019-12)
YOLACT++: Better Real-time Instance Segmentation
2004.10934 (2020-04)
YOLOv4: Optimal Speed and Accuracy of Object Detection
2304.08069 (2023-04)
DETRs Beat YOLOs on Real-time Object Detection
and 1 more not named here
gen_encoder_output_proposals 4 other papers
2 dated earlier, 2 later
2203.03605 (2022-03)
DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection
2206.02777 (2022-06)
Mask DINO: Towards A Unified Transformer-based Framework for Object Detection and Segmentation
2310.08530 (2023-10)
X-Pose: Detecting Any Keypoints
and 1 more not named here
dict_merge 3 other papers
2 dated earlier, 1 later
2004.10934 (2020-04)
YOLOv4: Optimal Speed and Accuracy of Object Detection
2304.08069 (2023-04)
DETRs Beat YOLOs on Real-time Object Detection
2403.00326 (2024-03)
DAMSDet: Dynamic Adaptive Multispectral Detection Transformer with Competitive Query Selection and Adaptive Feature Fusion
load_config 3 other papers
2 dated earlier, 1 later
2004.10934 (2020-04)
YOLOv4: Optimal Speed and Accuracy of Object Detection
2304.08069 (2023-04)
DETRs Beat YOLOs on Real-time Object Detection
2403.00326 (2024-03)
DAMSDet: Dynamic Adaptive Multispectral Detection Transformer with Competitive Query Selection and Adaptive Feature Fusion
get_weights_url 2 other papers
1 dated earlier, 1 later
2004.10934 (2020-04)
YOLOv4: Optimal Speed and Accuracy of Object Detection
2403.00326 (2024-03)
DAMSDet: Dynamic Adaptive Multispectral Detection Transformer with Competitive Query Selection and Adaptive Feature Fusion
get_affine_mat_kernel 1 other paper
1 dated earlier
2004.10934 (2020-04)
YOLOv4: Optimal Speed and Accuracy of Object Detection
get_affine_transform 1 other paper
1 dated earlier
2004.10934 (2020-04)
YOLOv4: Optimal Speed and Accuracy of Object Detection
get_warp_matrix 1 other paper
1 dated earlier
2004.10934 (2020-04)
YOLOv4: Optimal Speed and Accuracy of Object Detection
multiclass_nms 1 other paper
1 later
2403.00326 (2024-03)
DAMSDet: Dynamic Adaptive Multispectral Detection Transformer with Competitive Query Selection and Adaptive Feature Fusion
nms 1 other paper
1 later
2403.00326 (2024-03)
DAMSDet: Dynamic Adaptive Multispectral Detection Transformer with Competitive Query Selection and Adaptive Feature Fusion
norm_angle 1 other paper
1 later
2403.00326 (2024-03)
DAMSDet: Dynamic Adaptive Multispectral Detection Transformer with Competitive Query Selection and Adaptive Feature Fusion
poly2rbox_le135_np 1 other paper
1 later
2403.00326 (2024-03)
DAMSDet: Dynamic Adaptive Multispectral Detection Transformer with Competitive Query Selection and Adaptive Feature Fusion
poly2rbox_oc_np 1 other paper
1 later
2403.00326 (2024-03)
DAMSDet: Dynamic Adaptive Multispectral Detection Transformer with Competitive Query Selection and Adaptive Feature Fusion

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

In this paper, we study the problem of end-to-end multiperson pose estimation. State-of-the-art solutions adopt the DETR-like framework, and mainly develop the complex decoder, e.g., regarding pose estimation as keypoint box detection and combining with human detection in ED-Pose [38], hierarchically predicting with pose decoder and joint (keypoint) decoder in PETR [27]. We present a simple yet effective transformer approach, named Group Pose. We simply regard K-keypoint pose estimation as predicting a set of N × K keypoint positions, each from a keypoint query, as well as representing each pose with an instance query for scoring N pose predictions. Motivated by the intuition that the interaction, among across-instance queries of different types, is not directly helpful, we make a simple modification to decoder selfattention. We replace single self-attention over all the N × (K + 1) queries with two subsequent group selfattentions: (i) N within-instance self-attention, with each over K keypoint queries and one instance query, and (ii) (K +1) same-type across-instance self-attention, each over N queries of the same type. The resulting decoder removes the interaction among across-instance type-different queries, easing the optimization and thus improving the performance. Experimental results on MS COCO and Crowd-Pose show that our approach without human box supervision is superior to previous methods with complex decoders, and even is slightly better than ED-Pose that uses human box supervision. Paddle 1 and PyTorch 2 codes are available.

For agents

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

get_harvested_code_for_paper("2308.07313")
get_code_for_paper("2308.07313")
have("2308.07313")

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

curl https://syntology.ai/api/ran/2308.07313.json

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

[![Syntology run record](https://syntology.ai/api/ran/2308.07313.svg)](https://syntology.ai/paper/2308.07313)

Connect an agent — have() is free.