SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2306.02850 · CVPR · 2023

TRACE: 5D Temporal Regression of Avatars with Dynamic Cameras in 3D Environments

Yu Sun, Tao Mei, Michael Black, Wu Liu, Qian Bao

arXiv · PDF · Open in the Atlas

Code that ran

15 of 20 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 10 have no contract verdict; 4 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/2306.02850.json · badge: /api/ran/2306.02850.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
arthur151/dynacam canonical (classifier label)
classifier basis: census
7 ran · 0 no recorded run
Arthur151/ROMP canonical (classifier label)
classifier basis: census
5 ran · 4 no recorded run
Arthur151/CenterHMR community (classifier label)
classifier basis: plan_row
3 ran · 1 no recorded run
FunctionStatusWhere it lives
BHWC_to_BCHW Ran
ran on a synthesized input
Arthur151/ROMP/simple_romp/romp/model.py
code served (permissive licence) · get_code("eeaac5bd5e411a86")
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, 5, 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)

# The function expects a torch tensor with shape B x H x W x C
# Let's create a small tensor with batch=2, height=4, width=3, channels=5
x = torch.randn(2, 4, 3, 5)

ARGS = {
    "x": x
}
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, 5, 4, 3) — shape (2, 5, 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
angle_axis_to_rotation_matrix Ran
ran on a synthesized input
arthur151/dynacam/evaluation/utils.py
code served (permissive licence) · get_code("6e7ef9c3ecc8c368")
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, 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 numpy as np

torch.manual_seed(0)

# The function expects:
# - aas: a tensor with a .numpy() method
# The tensor is indexed in a loop, so it should be 2D (batch_size, 3)
# where each row is a 3-element angle-axis vector (cv2.Rodrigues expects 3D vectors)

aas = torch.randn(2, 3).float()

ARGS = {
    'aas': aas
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float, shape (2, 3, 3) — shape (2, 3, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
cam2pixel Ran
ran, honoured the contract we drafted
Arthur151/ROMP/romp/lib/dataset/MuPoTS.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from xuanhuang0/vanerf, is pointer only (licence: NONE) · get_code("5bbf1999f902db71")
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 xuanhuang0/vanerf, 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
Arthur151/ROMP/simple_romp/romp/model.py
pointer only (licence: Apache-2.0); 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 (Apache-2.0), 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_1D Ran
ran; our contract draft was wrong, not the code
Arthur151/CenterHMR/simple_romp/bev/model.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from jaewoo97/t2p, is pointer only (licence: NONE) · get_code("3423d9c4e59dae15")
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 jaewoo97/t2p, 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_3D Ran
ran; our contract draft was wrong, not the code
Arthur151/CenterHMR/simple_romp/bev/model.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from jaewoo97/t2p, is pointer only (licence: NONE) · get_code("e144d4d060a34395")
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 jaewoo97/t2p, 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.

get_3Dcoord_maps_halfz Ran
ran; our contract draft was wrong, not the code
Arthur151/CenterHMR/simple_romp/bev/model.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from jaewoo97/t2p, is pointer only (licence: NONE) · get_code("367e379193fd745f")
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 jaewoo97/t2p, 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.

get_cam3dmap_anchor Ran
ran on a synthesized input
Arthur151/ROMP/romp/lib/constants.py
code served (permissive licence) · get_code("7d9c9ae5d8d23f0e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 1, dtype family float, shape (32,) 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

ARGS = {
    "FOV": 45.0,
    "centermap_size": 64
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (32,) — shape (32,) 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_coord_maps Ran
ran on a synthesized input
Arthur151/ROMP/simple_romp/romp/model.py
code served (permissive licence) · get_code("b77632bd876028e5")
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 (1, 2, 128, 128) 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)

ARGS = {
    "size": 128
}
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 (1, 2, 128, 128) — shape (1, 2, 128, 128) 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
load_glamr_seq_results Ran
ran on a synthesized input
arthur151/dynacam/evaluation/loading_data.py
code served (permissive licence) · get_code("2067c2c16015d0d6")
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 os
import pickle
import tempfile
import numpy as np

# Create a temporary directory for results
temp_dir = tempfile.mkdtemp()

# Create mock GLAMR results data
person_data = {
    'person_0': {
        'frame2ind': {0: 0, 1: 1, 2: 2},
        'kp_2d_pred': np.random.rand(3, 17, 2),  # 3 frames, 17 keypoints, 2D coords
        'root_trans_world': np.random.rand(3, 3),  # 3 frames, 3D translation
        'smpl_orient_world': np.random.rand(3, 3, 3),  # 3 frames, 3x3 rotation matrix
    },
    'person_1': {
        'frame2ind': {0: 0, 1: 1},
        'kp_2d_pred': np.random.rand(2, 17, 2),  # 2 frames, 17 keypoints, 2D coords
        'root_trans_world': np.random.rand(2, 3),  # 2 frames, 3D translation
        'smpl_orient_world': np.random.rand(2, 3, 3),  # 2 frames, 3x3 rotation matrix
    }
}

results = {
    'person_data': person_data
}

# Save to pickle file with the expected naming convention
folder_name = 'test_sequence'
results_folder = temp_dir
output_file = os.path.join(results_folder, folder_name + '_seed1.pkl')

with open(output_file, 'wb') as f:
    pickle.dump(results, f)

ARGS = {
    'folder': folder_name,
    'results_folder': results_folder,
}
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
load_gts Ran
ran on a synthesized input
arthur151/dynacam/evaluation/loading_data.py
code served (permissive licence) · get_code("be5b762cce53110f")
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 tempfile
import os

# Create a temporary directory and a mock .npz file
temp_dir = tempfile.mkdtemp()
npz_path = os.path.join(temp_dir, "world_annots.npz")

# Create mock annotation data
annots_dict = {
    'sequence_dict': {'seq1': [1, 2, 3]},
    'ID_num': 42,
    'seq_name_1': np.array([1.0, 2.0, 3.0]),
    'seq_name_2': np.array([4.0, 5.0, 6.0]),
}

# Save to .npz file
np.savez(npz_path, annots=annots_dict)

ARGS = {
    'world_annots_path': npz_path
}
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 — 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
match_kp2ds Ran
ran on a synthesized input
arthur151/dynacam/evaluation.py
code served (permissive licence) · get_code("5f0bf8984d756328")
How it ran

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

import numpy as np

# Analyze the function to understand what it expects:
# - preds: array indexed with [:, valid_mask], shape should be (num_preds, num_keypoints)
# - gts: iterable of gt arrays, each gt has shape (num_keypoints, feature_dim)
#   where gt[:, 0] is used to create a valid_mask
#   and gt[valid_mask] is used in broadcasting with preds[:, valid_mask]

# Create sample data
num_preds = 3
num_keypoints = 5
feature_dim = 2

# preds shape: (num_preds, num_keypoints, feature_dim)
# This allows preds[:, valid_mask] to work where valid_mask indexes keypoints
preds = np.random.randn(num_preds, num_keypoints, feature_dim)

# gts: list of gt arrays
# Each gt shape: (num_keypoints, feature_dim)
# gt[:, 0] gives first coordinate of each keypoint (used for valid_mask)
# gt[valid_mask] filters to valid keypoints
gt1 = np.random.randn(num_keypoints, feature_dim)
gt1[0, 0] = 1.0  # ensure at least one valid keypoint
gt1[2, 0] = -0.5  # invalid keypoint (<=0)

gt2 = np.random.randn(num_keypoints, feature_dim)
gt2[1, 0] = 2.0  # valid
gt2[3, 0] = 0.0  # invalid

gts = [gt1, gt2]

ARGS = {
    "preds": preds,
    "gts": gts,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family int, shape (2,) — shape (2,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
obtain_smpl_verts Ran
ran on a synthesized input
arthur151/dynacam/show_examples.py
code served (permissive licence) · get_code("a5fa84c8fadc2688")
How it ran

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

import torch
import torch.nn as nn

torch.manual_seed(0)

class MockSMPLModel(nn.Module):
    """Minimal SMPL model that returns joints tensor"""
    def __init__(self):
        super().__init__()
        # Register a dummy parameter so the module has learnable params
        self.register_parameter('dummy', nn.Parameter(torch.zeros(1)))
    
    def forward(self, global_orient, body_pose, betas):
        batch_size = global_orient.shape[0]
        # Return an object with a joints attribute
        # global_orient: (batch, 3), body_pose: (batch, 69), betas: (batch, 10)
        # joints output shape: (batch, num_joints, 3)
        joints = torch.randn(batch_size, 23, 3)
        
        class SMPLOutput:
            def __init__(self, joints_tensor):
                self.joints = joints_tensor
        
        return SMPLOutput(joints)

# Create test data
batch_size = 2
smpl_thetas = torch.randn(batch_size, 73)  # 1 global_orient (3,) + 24 body joints (72,)
smpl_betas = torch.randn(batch_size, 10)   # Shape parameter
smpl_model = MockSMPLModel()

ARGS = {
    'smpl_thetas': smpl_thetas,
    'smpl_betas': smpl_betas,
    'smpl_model': smpl_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 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
process_idx Ran
ran on a synthesized input
arthur151/dynacam/evaluation/loading_data.py
code served (permissive licence) · get_code("632285e7a9c7aa01")
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

# Create a valid reorganize_idx array
# This should be a 1D array of indices that reference original image indices
reorganize_idx = np.array([0, 1, 0, 2, 1, 0, 2, 2], dtype=np.int64)

# vids is optional and not used in the function body
vids = None

ARGS = {
    "reorganize_idx": reorganize_idx,
    "vids": vids,
}
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 — 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
rotation_matrix_to_angle_axis Ran
ran on a synthesized input
arthur151/dynacam/evaluation/utils.py
code served (permissive licence) · get_code("c7fbe17ed134d674")
How it ran

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

import torch
import numpy as np
import cv2

torch.manual_seed(0)

# The function expects rotmats to be a torch tensor with shape (batch_size, 3, 3)
# Each element is a 3x3 rotation matrix
# cv2.Rodrigues expects a 3x3 matrix and returns (angle_axis, jacobian)
# angle_axis has shape (3, 1), so we extract [:, 0] to get shape (3,)

batch_size = 2
rotation_matrices = torch.randn(batch_size, 3, 3).float()

# Make them approximately valid rotation matrices by normalizing
# (not strictly orthogonal, but cv2.Rodrigues will process them)
for i in range(batch_size):
    u, s, v = torch.linalg.svd(rotation_matrices[i])
    rotation_matrices[i] = u @ v

ARGS = {
    "rotmats": rotation_matrices
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 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
joint_mapping Not yet run
unverified (harvested, no recorded run)
Arthur151/ROMP/romp/lib/constants.py
code served (permissive licence) · get_code("3d8f597016be8b2c")
How it ran

no invocation recorded.

make_cuda_ext Not yet run
unverified (harvested, no recorded run)
Arthur151/ROMP/simple_romp/setup_trace.py
code served (permissive licence) · get_code("036ffacc924ae8c3")
How it ran

no invocation recorded.

parse_args Not yet run
unverified (harvested, no recorded run)
Arthur151/ROMP/romp/lib/config.py
code served (permissive licence) · get_code("de251ea161b15c06")
How it ran

no invocation recorded.

parse_args Not yet run
unverified (harvested, no recorded run)
Arthur151/CenterHMR/trace/lib/config.py
code served (permissive licence) · get_code("91fa9a77ff72a999")
How it ran

no invocation recorded.

read_calibration Not yet run
unverified (harvested, no recorded run)
Arthur151/ROMP/romp/lib/dataset/MuCo.py
code served (permissive licence) · get_code("21c336c134a4f826")
How it ran

no invocation recorded.

Also present in other papers' code

Of 20 samples lifted from this paper's linked repositories, 20 carry a function identity and 0 do not. Of the 20 distinct functions with an identity, 8 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 1 the same month, 5 later) and 12 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

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

FunctionAlso present inEarliest-datedOthers named
conv3x3 686 other papers
482 dated earlier, 6 the same month, 191 later, 7 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
cam2pixel 3 other papers
2 dated earlier, 1 later
1907.11346 (2019-07)
Camera Distance-aware Top-down Approach for 3D Multi-person Pose Estimation from a Single RGB Image
2112.08274 (2021-12)
Putting People in their Place: Monocular Regression of 3D People in Depth
2401.00979 (2024-01)
3D Visibility-Aware Generalizable Neural Radiance Fields for Interacting Hands
BHWC_to_BCHW 1 other paper
1 dated earlier
2112.08274 (2021-12)
Putting People in their Place: Monocular Regression of 3D People in Depth
conv3x3_1D 1 other paper
1 later
2404.05218 (2024-04)
Multi-agent Long-term 3D Human Pose Forecasting via Interaction-aware Trajectory Conditioning
conv3x3_3D 1 other paper
1 later
2404.05218 (2024-04)
Multi-agent Long-term 3D Human Pose Forecasting via Interaction-aware Trajectory Conditioning
get_3Dcoord_maps_halfz 1 other paper
1 later
2404.05218 (2024-04)
Multi-agent Long-term 3D Human Pose Forecasting via Interaction-aware Trajectory Conditioning
get_coord_maps 1 other paper
1 dated earlier
2303.05938 (2023-03)
ACR: Attention Collaboration-based Regressor for Arbitrary Two-Hand Reconstruction
read_calibration 1 other paper
1 dated earlier
2109.02303 (2021-09)
Encoder-decoder with Multi-level Attention for 3D Human Shape and Pose Estimation

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

Figure 1. 5D temporal reconstruction of multiple 3D people in global coordinates with a dynamic camera. We introduce TRACE, a monocular one-stage method with a holistic 5D representation that enables the network to explicitly reason about human motion across time. Unlike previous methods [44,45] (b) that regress all 3D people from each frame in camera coordinates, TRACE (c) tracks the subjects (a) presented in the first frame through time and recovers their global trajectories in global coordinates in one shot.

For agents

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

get_harvested_code_for_paper("2306.02850")
get_code_for_paper("2306.02850")
have("2306.02850")

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

curl https://syntology.ai/api/ran/2306.02850.json

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

[![Syntology run record](https://syntology.ai/api/ran/2306.02850.svg)](https://syntology.ai/paper/2306.02850)

Connect an agent — have() is free.