SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2107.13774 · ICCV · 2021

Geometry Uncertainty Projection Network for Monocular 3D Object Detection

Junjie Yan, Wanli Ouyang, Yan Lu, Xinzhu Ma, Lei Yang, Qi Chu, Tianzhu Zhang, Yating Liu

arXiv · PDF · Open in the Atlas

Code that ran

16 of 21 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, 1 violated it and 6 have no contract verdict; 8 failures of our instrument; 5 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/2107.13774.json · badge: /api/ran/2107.13774.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
supermhp/gupnet canonical (classifier label)
classifier basis: census
12 ran · 2 no recorded run
SuperMHP/GUPNet community
Papers with Code archive
4 ran · 3 no recorded run
FunctionStatusWhere it lives
_gather_feat Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
SuperMHP/GUPNet/code/lib/models/gupnet.py
code served (permissive licence) · get_code("96f6547d6d3a3fce")
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 (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)

# Construct tensors for _gather_feat
# feat: B * (H*W) * C
# B=2 (batch), H*W=6 (spatial), C=4 (channels)
feat = torch.randn(2, 6, 4)

# ind: B * K where K=3 (number of indices to gather)
# Values should be valid indices into dimension 1 (0-5)
ind = torch.tensor([[0, 2, 4], [1, 3, 5]])

# mask: B * K with same shape as ind
# Boolean tensor for masking
mask = torch.tensor([[True, False, True], [True, True, False]])

ARGS = {
    "feat": feat,
    "ind": ind,
    "mask": mask,
}
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 (4, 4) — shape (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
_nms Ran
ran; our contract draft was wrong, not the code
SuperMHP/GUPNet/code/lib/models/gupnet.py
code served (permissive licence) · get_code("9cc3c87feb179eff")
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 (2, 1, 4, 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 torch
import torch.nn as nn

torch.manual_seed(0)

# Create a small heatmap tensor
# Shape: (batch=2, channels=1, height=4, width=3)
heatmap = torch.rand(2, 1, 4, 3)

# Create a kernel size parameter
kernel = 3

ARGS = {
    "heatmap": heatmap,
    "kernel": kernel,
}
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 (2, 1, 4, 3) — shape (2, 1, 4, 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
_transpose_and_gather_feat Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
SuperMHP/GUPNet/code/lib/models/gupnet.py
code served (permissive licence) · get_code("062a3f34484aa5ae")
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, 6, 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)

# Construct small tensors for the function
# feat: feature maps shaped in B * C * H * W
# Let's use B=2, C=4, H=3, W=5
batch_size = 2
channels = 4
height = 3
width = 5
k = 6  # number of indices

# Create feat tensor: B * C * H * W
feat = torch.randn(batch_size, channels, height, width)

# Create ind tensor: B * K
# Indices should be valid positions in the flattened H*W dimension
ind = torch.randint(0, height * width, (batch_size, k))

ARGS = {
    "feat": feat,
    "ind": ind,
}
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, 6, 4) — shape (2, 6, 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
angle2class Ran
ran, honoured the contract we drafted
supermhp/gupnet/code/lib/datasets/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from zhanyang-nwpu/mono3dvg, is pointer only (licence: NONE) · get_code("39dc7b17b3bf3218")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from zhanyang-nwpu/mono3dvg, 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.

build_optimizer Ran
ran on a synthesized input
supermhp/gupnet/code/lib/helpers/optimizer_helper.py
code served (permissive licence) · get_code("d7531c865af9dac9")
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 Adam. 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 torch.nn as nn
import torch.optim as optim

torch.manual_seed(0)

# Create a small model with parameters (weights and biases)
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.linear2(x)
        return x

model = SimpleModel()

# Create optimizer config
cfg_optimizer = {
    'type': 'adam',
    'lr': 0.001,
    'weight_decay': 0.0001
}

ARGS = {
    'cfg_optimizer': cfg_optimizer,
    'model': model
}
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 Adam — a Adam output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Adam 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
check_range Ran
ran on a synthesized input
supermhp/gupnet/code/lib/datasets/utils.py
code served (permissive licence) · get_code("e2619cecce6ca5d5")
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

def check_range(angle):
    if angle > np.pi:
        angle -= 2 * np.pi
    if angle < -np.pi:
        angle += 2 * np.pi
    return angle

# Test with a scalar angle value
angle_value = 4.0  # This is > np.pi, so it will be adjusted

ARGS = {
    "angle": angle_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
conv3x3 Ran
ran; our contract draft was wrong, not the code
supermhp/gupnet/code/lib/backbones/dla.py
pointer only (licence: MIT); the stored copy get_code serves, from lixiaotong97/DSU, is pointer only (licence: NONE) · get_code("fac5364e2f53c6db")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from lixiaotong97/DSU, 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.

conv3x3 Ran
ran; our contract draft was wrong, not the code
supermhp/gupnet/code/lib/backbones/resnet.py
code served (permissive licence) · get_code("c875390a9fab8007")
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 Conv2d. 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 torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "in_planes": 3,
    "out_planes": 64,
    "stride": 1,
}
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 Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d 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
extract_input_from_tensor Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
SuperMHP/GUPNet/code/lib/models/gupnet.py
code served (permissive licence) · get_code("f6907c844231eed4")
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 (8, 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 torch.nn as nn

torch.manual_seed(0)

# Build tiny tensors for the function
# extract_input_from_tensor expects:
# - input: feature maps shaped in B * C * H * W
# - ind: indices tensor shaped in B * K
# - mask: tensor shaped in B * K (boolean mask)

B = 2  # batch size
C = 4  # channels
H = 3  # height
W = 3  # width
K = 5  # number of indices/mask entries

# Create input tensor: B * C * H * W
input_tensor = torch.randn(B, C, H, W)

# Create ind tensor: B * K with valid indices for H*W = 9
# ind values should be in range [0, H*W-1] = [0, 8]
ind_tensor = torch.randint(0, H * W, (B, K))

# Create mask tensor: B * K (boolean)
mask_tensor = torch.ones(B, K, dtype=torch.bool)
# Set some values to False to test masking
mask_tensor[0, 2:4] = False

ARGS = {
    "input": input_tensor,
    "ind": ind_tensor,
    "mask": mask_tensor,
}
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 (8, 4) — shape (8, 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
get_angle_from_box3d Ran
ran on a synthesized input
supermhp/gupnet/code/lib/datasets/utils.py
code served (permissive licence) · get_code("ea47cdbdb36b7bc7")
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 float64, ndim 0, 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 expects box3d_pts_3d which is indexed as:
# - box3d_pts_3d[0], box3d_pts_3d[1], box3d_pts_3d[2], box3d_pts_3d[3]
# - Each element is indexed with [0] and [-1], so they're array-like with at least 2 elements
# Creating a list of 4 points, each with 3D coordinates (x, y, z)

box3d_pts_3d = [
    np.array([1.0, 2.0, 3.0]),      # point 0
    np.array([2.0, 3.0, 4.0]),      # point 1
    np.array([0.5, 1.5, 2.5]),      # point 2
    np.array([1.5, 2.5, 3.5])       # point 3
]

ARGS = {
    "box3d_pts_3d": box3d_pts_3d
}
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 float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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_calib_from_file Ran
ran on a synthesized input
supermhp/gupnet/code/lib/datasets/kitti_utils.py
code served (permissive licence) · get_code("ebd8a3fc2a270133")
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 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 calibration file with the expected format
temp_dir = tempfile.mkdtemp()
calib_file = os.path.join(temp_dir, "calib.txt")

with open(calib_file, 'w') as f:
    # Lines 0-1: dummy content
    f.write("P0: dummy\n")
    f.write("P1: dummy\n")
    # Line 2: P2 (3x4 = 12 values)
    f.write("P2: " + " ".join([str(i) for i in range(12)]) + "\n")
    # Line 3: P3 (3x4 = 12 values)
    f.write("P3: " + " ".join([str(i+12) for i in range(12)]) + "\n")
    # Line 4: R0 (3x3 = 9 values)
    f.write("R0: " + " ".join([str(i+24) for i in range(9)]) + "\n")
    # Line 5: Tr_velo_to_cam (3x4 = 12 values)
    f.write("Tr_velo_to_cam: " + " ".join([str(i+33) for i in range(12)]) + "\n")

ARGS = {
    "calib_file": calib_file
}
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 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
get_dir Ran
ran, violated the contract we drafted
supermhp/gupnet/code/lib/datasets/kitti_utils.py
code served (permissive licence) · get_code("5d546e1ecbee1866")
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

# Analyze the function:
# - src_point: used with indexing [0] and [1], so it needs to be indexable with 2 elements
# - rot_rad: used with np.sin() and np.cos(), so it should be a scalar number

# Create test arguments
src_point = [1.0, 2.0]  # A simple 2D point
rot_rad = np.pi / 4     # A rotation angle in radians

ARGS = {
    "src_point": src_point,
    "rot_rad": rot_rad
}
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
get_model_url Ran
ran; our contract draft was wrong, not the code
supermhp/gupnet/code/lib/backbones/dla.py
code served (permissive licence) · get_code("3c9867a84d7c18d9")
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

ARGS = {
    'data': 'imagenet',
    'name': 'dla34',
    'hash': 'ba72cf86'
}
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
get_objects_from_label Ran
ran; our contract draft was wrong, not the code
supermhp/gupnet/code/lib/datasets/kitti_utils.py
code served (permissive licence) · get_code("8274533a00c7a30a")
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 tempfile
import os

# Create a temporary label file with valid KITTI format lines
temp_dir = tempfile.mkdtemp()
label_file = os.path.join(temp_dir, 'test_labels.txt')

# Write valid Object3d initialization lines
# Format: type truncation occlusion alpha bbox(4) dimensions(3) location(3) rotation_y [score]
with open(label_file, 'w') as f:
    # Example line with score (16 fields)
    f.write('Car 0.0 0 0.5 100.0 150.0 200.0 250.0 1.5 1.6 3.5 10.0 5.0 20.0 0.785 0.95\n')
    # Example line without score (15 fields)
    f.write('Pedestrian 0.1 1 1.2 50.0 60.0 150.0 180.0 1.8 0.5 0.6 5.0 2.0 15.0 1.57\n')
    # DontCare example
    f.write('DontCare -1 -1 0.0 0.0 0.0 400.0 450.0 1.0 1.0 1.0 50.0 0.0 30.0 0.0\n')

ARGS = {
    'label_file': label_file
}
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
resnet18 Ran
ran on a synthesized input
supermhp/gupnet/code/lib/backbones/resnet.py
code served (permissive licence) · get_code("a2df86776904df7a")
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 ResNet. 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 torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'pretrained': False,
}
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 ResNet — a ResNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet 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
resnet34 Ran
ran on a synthesized input
supermhp/gupnet/code/lib/backbones/resnet.py
code served (permissive licence) · get_code("03238865f53f8f40")
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 ResNet. 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 torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'pretrained': False,
}
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 ResNet — a ResNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet 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
GUPNet Not yet run
unverified (harvested, no recorded run)
SuperMHP/GUPNet/code/lib/models/gupnet.py
code served (permissive licence) · get_code("0e9f041e43897d96")
How it ran

no invocation recorded.

_topk Not yet run
unverified (harvested, no recorded run)
SuperMHP/GUPNet/code/lib/models/gupnet.py
code served (permissive licence) · get_code("026b32da103167e6")
How it ran

no invocation recorded.

dla34 Not yet run
unverified (harvested, no recorded run)
supermhp/gupnet/code/lib/backbones/dla.py
code served (permissive licence) · get_code("1e380a5690a619ef")
How it ran

no invocation recorded.

extract_dets_from_outputs Not yet run
unverified (harvested, no recorded run)
supermhp/gupnet/code/lib/helpers/decode_helper.py
code served (permissive licence) · get_code("615edfe53f0e9355")
How it ran

no invocation recorded.

weights_init_xavier Not yet run
unverified (harvested, no recorded run)
SuperMHP/GUPNet/code/lib/models/gupnet.py
code served (permissive licence) · get_code("cb4724b9596507d8")
How it ran

no invocation recorded.

Also present in other papers' code

Of 21 samples lifted from this paper's linked repositories, 21 carry a function identity and 0 do not. Of the 21 distinct functions with an identity, 15 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 1 the same month, 15 later) and 6 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
conv3x3 686 other papers
325 dated earlier, 3 the same month, 352 later, 6 same year, order not determinable · 31 only as an assembly brick
1406.2661 (2014-06)
Generative Adversarial Networks
1409.0575 (2014-09)
ImageNet Large Scale Visual Recognition Challenge
1412.6572 (2014-12)
Explaining and Harnessing Adversarial Examples
and 683 more not named here
conv3x3 7 other papers
7 later
2205.13462 (2022-05)
FedBR: Improving Federated Learning on Heterogeneous Data via Local Learning Bias Reduction
2207.08531 (2022-07)
DID-M3D: Decoupling Instance Depth for Monocular 3D Object Detection
2212.02191 (2022-12)
On the effectiveness of partial variance reduction in federated learning with heterogeneous data
and 4 more not named here
get_dir 7 other papers
3 dated earlier, 4 later · 1 only as an assembly brick
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
1908.07919 (2019-08)
Deep High-Resolution Representation Learning for Visual Recognition
2105.04154 (2021-05)
Unsupervised Human Pose Estimation through Transforming Shape Templates
and 4 more not named here
get_calib_from_file 5 other papers
4 dated earlier, 1 later
1812.04244 (2018-12)
PointRCNN: 3D Object Proposal Generation and Detection from Point Cloud
2006.11275 (2020-06)
Center-based 3D Object Detection and Tracking
2007.08856 (2020-07)
EPNet: Enhancing Point Features with Image Semantics for 3D Object Detection
and 2 more not named here
angle2class 4 other papers
4 later
2207.08531 (2022-07)
DID-M3D: Decoupling Instance Depth for Monocular 3D Object Detection
2312.08022 (2023-12)
Mono3DVG: 3D Visual Grounding in Monocular Images
and 1 more not named here
get_model_url 4 other papers
4 later
2112.04628 (2021-12)
Learning Auxiliary Monocular Contexts Helps Monocular 3D Object Detection
2203.10981 (2022-03)
MonoDTR: Monocular 3D Object Detection with Depth-Aware Transformer
2207.08531 (2022-07)
DID-M3D: Decoupling Instance Depth for Monocular 3D Object Detection
and 1 more not named here
_transpose_and_gather_feat
brick of this paper's assembly: helper
3 other papers
1 dated earlier, 2 later · 1 only as an assembly brick
2104.02323 (2021-04)
Objects are Different: Flexible Monocular 3D Object Detection
2108.11127 (2021-08)
AutoShape: Real-Time Shape-Aware Monocular 3D Object Detection
2304.02173 (2023-04)
ChartReader: A Unified Framework for Chart Derendering and Comprehension without Heuristic Rules
get_objects_from_label 3 other papers
1 dated earlier, 2 later
2006.11275 (2020-06)
Center-based 3D Object Detection and Tracking
2207.08531 (2022-07)
DID-M3D: Decoupling Instance Depth for Monocular 3D Object Detection
2310.07602 (2023-10)
Dual Radar: A Multi-modal Dataset with Dual 4D Radar for Autonomous Driving
_gather_feat
brick of this paper's assembly: helper
2 other papers
1 dated earlier, 1 later · 1 only as an assembly brick
2104.02323 (2021-04)
Objects are Different: Flexible Monocular 3D Object Detection
2108.11127 (2021-08)
AutoShape: Real-Time Shape-Aware Monocular 3D Object Detection
build_optimizer 1 other paper
1 later
2207.08531 (2022-07)
DID-M3D: Decoupling Instance Depth for Monocular 3D Object Detection
check_range 1 other paper
1 later
2207.08531 (2022-07)
DID-M3D: Decoupling Instance Depth for Monocular 3D Object Detection
dla34 1 other paper
1 later
2203.10981 (2022-03)
MonoDTR: Monocular 3D Object Detection with Depth-Aware Transformer
get_angle_from_box3d 1 other paper
1 later
2207.08531 (2022-07)
DID-M3D: Decoupling Instance Depth for Monocular 3D Object Detection
resnet18 1 other paper
1 later
2207.08531 (2022-07)
DID-M3D: Decoupling Instance Depth for Monocular 3D Object Detection
resnet34 1 other paper
1 later
2207.08531 (2022-07)
DID-M3D: Decoupling Instance Depth for Monocular 3D Object Detection

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

Geometry Projection is a powerful depth estimation method in monocular 3D object detection. It estimates depth dependent on heights, which introduces mathematical priors into the deep model. But projection process also introduces the error amplification problem, in which the error of the estimated height will be amplified and reflected greatly at the output depth. This property leads to uncontrollable depth inferences and also damages the training efficiency. In this paper, we propose a Geometry Uncertainty Projection Network (GUP Net) to tackle the error amplification problem at both inference and training stages. Specifically, a GUP module is proposed to obtains the geometryguided uncertainty of the inferred depth, which not only provides high reliable confidence for each depth but also benefits depth learning. Furthermore, at the training stage, we propose a Hierarchical Task Learning strategy to reduce the instability caused by error amplification. This learning algorithm monitors the learning situation of each task by a proposed indicator and adaptively assigns the proper loss weights for different tasks according to their pre-tasks situation. Based on that, each task starts learning only when its pre-tasks are learned well, which can significantly improve the stability and efficiency of the training process. Extensive experiments demonstrate the effectiveness of the proposed method. The overall model can infer more reliable object depth than existing methods and outperforms the state-of-the-art image-based monocular 3D detectors by 3.74% and 4.7% AP 40 of the car and pedestrian categories on the KITTI benchmark. The code and model will be released at https://github.com/SuperMHP/GUPNet. † This work was done when Yan Lu was an intern at SenseTime.

For agents

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

get_harvested_code_for_paper("2107.13774")
get_code_for_paper("2107.13774")
have("2107.13774")

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

curl https://syntology.ai/api/ran/2107.13774.json

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

[![Syntology run record](https://syntology.ai/api/ran/2107.13774.svg)](https://syntology.ai/paper/2107.13774)

Connect an agent — have() is free.