SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2312.03641 · 2023

MotionCtrl: A Unified and Flexible Motion Controller for Video Generation

arXiv · PDF · Open in the Atlas

Code that ran

17 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; 2 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 14 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/2312.03641.json · badge: /api/ran/2312.03641.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
TencentARC/MotionCtrl canonical (classifier label)
classifier basis: census
17 ran · 5 no recorded run
FunctionStatusWhere it lives
autocast Ran
ran; our contract draft was wrong, not the code
TencentARC/MotionCtrl/lvdm/common.py
code served (permissive licence) · get_code("32af1c12fd826ef2")
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 function. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type 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 autocast function takes a single parameter: a function f
# It returns a wrapped version of that function with autocast enabled

# Create a simple function to wrap
def sample_forward(x):
    """A simple function that does some computation"""
    return x * 2 + 1

ARGS = {
    "f": sample_forward
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
calc_cam_cone_pts_3d Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
TencentARC/MotionCtrl/gradio_utils/utils.py
code served (permissive licence) · get_code("0431aefa5f6f913f")
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 (5, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Construct valid arguments for calc_cam_cone_pts_3d

# R_W2C: rotation matrix from world to camera (3x3)
R_W2C = np.array([
    [1.0, 0.0, 0.0],
    [0.0, 1.0, 0.0],
    [0.0, 0.0, 1.0]
], dtype=np.float32)

# T_W2C: translation vector from world to camera (3,)
T_W2C = np.array([0.1, 0.2, 0.3], dtype=np.float32)

# fov_deg: field of view in degrees (scalar)
fov_deg = 45.0

# scale: scale factor for cone points (scalar)
scale = 0.1

# set_canonical: whether to use canonical frame (boolean)
set_canonical = False

# first_frame_RT: first frame rotation-translation matrix (3x4) - only used if set_canonical=True
first_frame_RT = None

ARGS = {
    "R_W2C": R_W2C,
    "T_W2C": T_W2C,
    "fov_deg": fov_deg,
    "scale": scale,
    "set_canonical": set_canonical,
    "first_frame_RT": first_frame_RT
}
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 (5, 3) — shape (5, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
compute_R_form_rad_angle Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
TencentARC/MotionCtrl/gradio_utils/camera_utils.py
code served (permissive licence) · get_code("4981c629ab3defc9")
How it ran

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

import numpy as np

# The function takes angles as a parameter
# It unpacks angles into three components: theta_x, theta_y, theta_z
# These are used in trigonometric functions, so they should be numeric values (floats or arrays)
# The function constructs rotation matrices using these angles

# Create sample angles in radians
angles = (np.pi / 6, np.pi / 4, np.pi / 3)  # 30°, 45°, 60° in radians

ARGS = {
    "angles": angles
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (3, 3) — shape (3, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
create_relative Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
TencentARC/MotionCtrl/gradio_utils/camera_utils.py
code served (permissive licence) · get_code("ba8074536bf63264")
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
import copy

np.random.seed(0)

# RT_list: list of rotation-translation matrices
# The function expects each element to be a 3x4 matrix (3 rows for 3D, 4 columns for R and T)
# Create two sample RT matrices
RT_1 = np.array([
    [1.0, 0.0, 0.0, 1.0],
    [0.0, 1.0, 0.0, 2.0],
    [0.0, 0.0, 1.0, 3.0]
], dtype=np.float32)

RT_2 = np.array([
    [0.9, -0.1, 0.0, 1.5],
    [0.1, 0.9, 0.0, 2.5],
    [0.0, 0.0, 1.0, 3.5]
], dtype=np.float32)

RT_list = [copy.deepcopy(RT_1), copy.deepcopy(RT_2)]

ARGS = {
    'RT_list': RT_list,
    'K_1': 4.7,
    'dataset': 'syn'
}
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
disabled_train Ran
ran, violated the contract we drafted
TencentARC/MotionCtrl/lvdm/basics.py
code served (permissive licence) · get_code("4cb732f513d69dfd")
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 SimpleModel. 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 SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)

model = SimpleModel()

ARGS = {
    "self": model,
    "mode": True
}
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 SimpleModel — a SimpleModel output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SimpleModel 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_into_tensor Ran
ran; our contract draft was wrong, not the code
TencentARC/MotionCtrl/lvdm/common.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from Luo-Yihong/YOSO, is pointer only (licence: NONE) · get_code("a15cfd7932844ef9")
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 Luo-Yihong/YOSO, 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.

load_camera_pose Ran
ran; our contract draft was wrong, not the code
TencentARC/MotionCtrl/main/evaluation/motionctrl_inference.py
code served (permissive licence) · get_code("801a42f99feaacb4")
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 json
import numpy as np
import torch
import tempfile
import os

torch.manual_seed(0)

# Create a temporary directory structure with camera pose JSON files
temp_dir = tempfile.mkdtemp()
cond_dir = temp_dir
poses_dir = os.path.join(cond_dir, 'camera_poses')
os.makedirs(poses_dir, exist_ok=True)

# Create sample camera pose JSON files
camera_pose_names = ['test_camera_pose_0', 'test_camera_pose_1']
for pose_name in camera_pose_names:
    pose_data = [[1.0, 0.0, 0.0, 0.5],
                 [0.0, 1.0, 0.0, 0.5],
                 [0.0, 0.0, 1.0, 1.0],
                 [0.0, 0.0, 0.0, 1.0]]
    pose_file_path = os.path.join(poses_dir, f'{pose_name}.json')
    with open(pose_file_path, 'w') as f:
        json.dump(pose_data, f)

ARGS = {
    'cond_dir': cond_dir,
    'camera_poses': camera_pose_names
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_model_checkpoint Ran
ran; our contract draft was wrong, not the code
TencentARC/MotionCtrl/main/evaluation/motionctrl_inference.py
code served (permissive licence) · get_code("e03150391a8a3fbc")
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 MinimalModel. 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
from collections import OrderedDict
import tempfile
import os

torch.manual_seed(0)

# Create a minimal model with an adapter attribute
class MinimalModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
        self.adapter = nn.Linear(3, 2)
    
    def forward(self, x):
        return self.linear(x)

# Create temporary checkpoint files
temp_dir = tempfile.mkdtemp()

# Create model checkpoint with 'state_dict' key
model_for_ckpt = MinimalModel()
model_state = model_for_ckpt.state_dict()
ckpt_path = os.path.join(temp_dir, "model.pt")
torch.save({'state_dict': model_state}, ckpt_path)

# Create adapter checkpoint with 'state_dict' key
adapter_state = model_for_ckpt.adapter.state_dict()
adapter_ckpt_path = os.path.join(temp_dir, "adapter.pt")
torch.save({'state_dict': adapter_state}, adapter_ckpt_path)

# Create the model instance to load into
model = MinimalModel()

ARGS = {
    'model': model,
    'ckpt': ckpt_path,
    'adapter_ckpt': adapter_ckpt_path
}
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 MinimalModel — a MinimalModel output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a MinimalModel 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_trajs Ran
ran; our contract draft was wrong, not the code
TencentARC/MotionCtrl/main/evaluation/motionctrl_inference.py
code served (permissive licence) · get_code("8079e85f7bad9b5e")
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 torch
import tempfile
import os

torch.manual_seed(0)

# Create a temporary directory structure with dummy trajectory files
temp_dir = tempfile.mkdtemp()
traj_dir = os.path.join(temp_dir, 'trajectories')
os.makedirs(traj_dir, exist_ok=True)

# Create small dummy trajectory numpy files
# The function expects shape (H, W, C, 3) -> after permute (3, H, W, C)
# Creating trajectories with shape (4, 5, 6, 3) so after permute we get (3, 4, 5, 6)
traj_names = ['traj_001', 'traj_002']
for traj_name in traj_names:
    dummy_data = np.random.randn(4, 5, 6, 3).astype(np.float32)
    np.save(os.path.join(traj_dir, f'{traj_name}.npy'), dummy_data)

ARGS = {
    'cond_dir': temp_dir,
    'trajs': traj_names
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
mesh_grid Ran
ran; our contract draft was wrong, not the code
TencentARC/MotionCtrl/gradio_utils/flow_utils.py
code served (permissive licence) · get_code("7b60c906307a7a95")
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

# kernel_size should be an integer representing the size of the kernel
# The function generates a mesh grid centered at zero with shape (kernel_size, kernel_size, 2)
kernel_size = 5

ARGS = {
    "kernel_size": kernel_size
}
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
normal_kl Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
TencentARC/MotionCtrl/lvdm/distributions.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from NeuralCarver/Michelangelo, is pointer only (licence: GPL-3.0) · get_code("17eb0a31c91b671c")
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 NeuralCarver/Michelangelo, and GPL-3.0, as GitHub detects it on that repository, is not in the permissive set Syntology serves inline. 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.

pdf2 Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
TencentARC/MotionCtrl/gradio_utils/flow_utils.py
code served (permissive licence) · get_code("1f5c7fd171926d1c")
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 (5, 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 a valid covariance matrix (2x2, symmetric positive definite)
sigma_matrix = np.array([
    [1.0, 0.1],
    [0.1, 1.0]
], dtype=np.float32)

# Create a grid with shape (K, K, 2) where K is kernel size
K = 5
x = np.linspace(-2, 2, K, dtype=np.float32)
y = np.linspace(-2, 2, K, dtype=np.float32)
xx, yy = np.meshgrid(x, y)
grid = np.stack([xx, yy], axis=2)  # Shape: (K, K, 2)

ARGS = {
    "sigma_matrix": sigma_matrix,
    "grid": grid
}
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 (5, 5) — shape (5, 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
process_points Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
TencentARC/MotionCtrl/gradio_utils/traj_utils.py
code served (permissive licence) · get_code("261e5e8d05f389c0")
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

# Analyze the function: process_points takes a 'points' parameter
# which should be a list of coordinate pairs [x, y]
# The function processes these points to return exactly 16 frames worth of points

# Test case 1: points list with fewer than 2 points (should return default)
points_short = [[100, 200]]

# Test case 2: points list with 2-15 points (should interpolate to 16)
points_medium = [[100, 100], [200, 200], [300, 150]]

# Test case 3: points list with 16+ points (should downsample to 16)
points_long = [[100 + i*10, 100 + i*5] for i in range(20)]

ARGS = {
    "points": points_medium
}
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
read_points Ran
ran, honoured the contract we drafted
TencentARC/MotionCtrl/gradio_utils/traj_utils.py
code served (permissive licence) · get_code("9e50653059623b26")
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 tempfile
import os

# Create a temporary file with point data
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt')
temp_file.write("10,20\n")
temp_file.write("30,40\n")
temp_file.write("50,60\n")
temp_file.write("70,80\n")
temp_file.write("90,100\n")
temp_file.close()

ARGS = {
    'file': temp_file.name,
    'video_len': 16,
    'reverse': False,
}
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 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
sigma_matrix2 Ran
ran, honoured the contract we drafted
TencentARC/MotionCtrl/gradio_utils/flow_utils.py
code served (permissive licence) · get_code("60be2f7d3411849d")
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, 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

# Define arguments for sigma_matrix2
sig_x = 1.5
sig_y = 2.3
theta = np.pi / 4  # 45 degrees in radians

ARGS = {
    "sig_x": sig_x,
    "sig_y": sig_y,
    "theta": theta
}
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, 2) — shape (2, 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
spatial_forward_BasicTransformerBlock Ran
ran; our contract draft was wrong, not the code
TencentARC/MotionCtrl/motionctrl/lvdm_modified_modules.py
code served (permissive licence) · get_code("26ff90b6372efbb0")
How it ran

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

import torch
import torch.nn as nn

torch.manual_seed(0)

class MockAttention(nn.Module):
    def __init__(self, dim, context_dim=None):
        super().__init__()
        self.dim = dim
        self.context_dim = context_dim or dim
        
    def forward(self, x, context=None, mask=None):
        return x

class MockNorm(nn.Module):
    def __init__(self, dim):
        super().__init__()
        self.weight = nn.Parameter(torch.ones(dim))
        
    def forward(self, x):
        return x

class MockFF(nn.Module):
    def __init__(self, dim):
        super().__init__()
        self.fc = nn.Linear(dim, dim)
        
    def forward(self, x):
        return self.fc(x)

class BasicTransformerBlock(nn.Module):
    def __init__(self, dim=4, context_dim=4, disable_self_attn=False):
        super().__init__()
        self.dim = dim
        self.context_dim = context_dim
        self.disable_self_attn = disable_self_attn
        
        self.norm1 = MockNorm(dim)
        self.norm2 = MockNorm(dim)
        self.norm3 = MockNorm(dim)
        
        self.attn1 = MockAttention(dim, context_dim)
        self.attn2 = MockAttention(dim, context_dim)
        
        self.ff = MockFF(dim)
        
    def forward(self, x, context=None, mask=None):
        if isinstance(context, dict):
            context = context['context']
        x = self.attn1(self.norm1(x), context=context if self.disable_self_attn else None, mask=mask) + x
        x = self.attn2(self.norm2(x), context=context, mask=mask) + x
        x = self.ff(self.norm3(x)) + x
        return x

# Create mock block instance
block = BasicTransformerBlock(dim=4, context_dim=3, disable_self_attn=False)

# Create test inputs
x = torch.randn(2, 4)  # batch=2, dim=4
context = torch.randn(2, 3)  # batch=2, context_dim=3
mask = torch.ones(2, 2)  # simple attention mask

ARGS = {
    'self': block,
    'x': x,
    'context': context,
    '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 (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
zero_module Ran
ran; our contract draft was wrong, not the code
TencentARC/MotionCtrl/lvdm/basics.py
code served (permissive licence) · get_code("129b804760b3115f")
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 SimpleModule. 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)

# Create a simple module with some parameters
class SimpleModule(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
        self.conv = nn.Conv1d(2, 2, 3)
    
    def forward(self, x):
        return x

module = SimpleModule()

# Verify the module has non-zero parameters before zeroing
for p in module.parameters():
    assert p.abs().sum() > 0, "Parameters should be non-zero initially"

ARGS = {
    "module": module
}
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 SimpleModule — a SimpleModule output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SimpleModule 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
TemporalTransformer_forward Not yet run
unverified (harvested, no recorded run)
TencentARC/MotionCtrl/motionctrl/lvdm_modified_modules.py
code served (permissive licence) · get_code("7c15eeec0b04c6c5")
How it ran

no invocation recorded.

gather_data Not yet run
unverified (harvested, no recorded run)
TencentARC/MotionCtrl/lvdm/common.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from shijianjian/stereocrafter-zero, is pointer only (licence: NONE) · get_code("b72d8eb89fd75383")
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 shijianjian/stereocrafter-zero, 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. No invocation recorded.

get_camera_motion Not yet run
unverified (harvested, no recorded run)
TencentARC/MotionCtrl/gradio_utils/camera_utils.py
code served (permissive licence) · get_code("0bb11816f4e72eb4")
How it ran

no invocation recorded.

get_provided_traj Not yet run
unverified (harvested, no recorded run)
TencentARC/MotionCtrl/gradio_utils/traj_utils.py
code served (permissive licence) · get_code("ab421bf522d31a63")
How it ran

no invocation recorded.

scale_module Not yet run
unverified (harvested, no recorded run)
TencentARC/MotionCtrl/lvdm/basics.py
code served (permissive licence) · get_code("9136b5f9b735334b")
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, 10 are also present in other papers whose code we hold (10 with at least one paper dated earlier, 7 the same month, 10 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
zero_module 196 other papers
81 dated earlier, 9 the same month, 105 later, 1 same year, order not determinable · 49 only as an assembly brick
2006.11239 (2020-06)
Denoising Diffusion Probabilistic Models
2010.02502 (2020-10)
Denoising Diffusion Implicit Models
2105.05233 (2021-05)
Diffusion Models Beat GANs on Image Synthesis
and 193 more not named here
disabled_train 173 other papers
64 dated earlier, 10 the same month, 99 later · 5 only as an assembly brick
2012.09841 (2020-12)
Taming Transformers for High-Resolution Image Synthesis
2102.05379 (2021-02)
Argmax Flows and Multinomial Diffusion: Learning Categorical Distributions
2104.07652 (2021-04)
Geometry-Free View Synthesis: Transformers and no 3D Priors
and 170 more not named here
normal_kl 34 other papers
11 dated earlier, 3 the same month, 19 later, 1 same year, order not determinable
2006.11239 (2020-06)
Denoising Diffusion Probabilistic Models
2209.15421 (2022-09)
TabDDPM: Modelling Tabular Data with Diffusion Models
2211.13221 (2022-11)
Latent Video Diffusion Models for High-Fidelity Long Video Generation
and 31 more not named here
extract_into_tensor 32 other papers
8 dated earlier, 3 the same month, 21 later · 7 only as an assembly brick
2006.11239 (2020-06)
Denoising Diffusion Probabilistic Models
2010.02502 (2020-10)
Denoising Diffusion Implicit Models
2112.02815 (2021-12)
Make It Move: Controllable Image-to-Video Generation with Text Descriptions
and 29 more not named here
autocast 14 other papers
3 dated earlier, 1 the same month, 10 later
2204.06125 (2022-04)
Hierarchical Text-Conditional Image Generation with CLIP Latents
2310.04687 (2023-10)
Targeted Attack Improves Protection against Unauthorized Diffusion Customization
2310.12190 (2023-10)
DynamiCrafter: Animating Open-domain Images with Video Diffusion Priors
and 11 more not named here
scale_module 14 other papers
3 dated earlier, 1 the same month, 10 later
2303.08863 (2023-03)
Class-Guided Image-to-Image Diffusion: Cell Painting from Brightfield Images with Class Labels
2310.12190 (2023-10)
DynamiCrafter: Animating Open-domain Images with Video Diffusion Priors
2310.15587 (2023-10)
SCANDL: A Diffusion Model for Generating Synthetic Scanpaths on Texts
and 11 more not named here
gather_data 10 other papers
1 dated earlier, 1 the same month, 8 later
2310.12190 (2023-10)
DynamiCrafter: Animating Open-domain Images with Video Diffusion Priors
2312.00330 (2023-12)
StyleCrafter: Enhancing Stylized Text-to-Video Generation with Style Adapter
2401.09047 (2024-01)
VideoCrafter2: Overcoming Data Limitations for High-Quality Video Diffusion Models
and 7 more not named here
mesh_grid 8 other papers
1 dated earlier, 7 later
2310.19011 (2023-10)
Efficient Test-Time Adaptation for Super-Resolution with Second-Order Degradation and Reconstruction
2403.05049 (2024-03)
XPSR: Cross-modal Priors for Diffusion-based Image Super-Resolution
2410.00418 (2024-10)
Posterior-Mean Rectified Flow: Towards Minimum MSE Photo-Realistic Image Restoration
and 5 more not named here
pdf2 8 other papers
1 dated earlier, 7 later
2310.19011 (2023-10)
Efficient Test-Time Adaptation for Super-Resolution with Second-Order Degradation and Reconstruction
2403.05049 (2024-03)
XPSR: Cross-modal Priors for Diffusion-based Image Super-Resolution
2410.00418 (2024-10)
Posterior-Mean Rectified Flow: Towards Minimum MSE Photo-Realistic Image Restoration
and 5 more not named here
sigma_matrix2 8 other papers
1 dated earlier, 7 later
2310.19011 (2023-10)
Efficient Test-Time Adaptation for Super-Resolution with Second-Order Degradation and Reconstruction
2403.05049 (2024-03)
XPSR: Cross-modal Priors for Diffusion-based Image Super-Resolution
2410.00418 (2024-10)
Posterior-Mean Rectified Flow: Towards Minimum MSE Photo-Realistic Image Restoration
and 5 more not named here

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

Repositories linked to this paper

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

Abstract

Motions in a video primarily consist of camera motion, induced by camera movement, and object motion, resulting from object movement. Accurate control of both camera and object motion is essential for video generation. However, existing works either mainly focus on one type of motion or do not clearly distinguish between the two, limiting their control capabilities and diversity. Therefore, this paper presents MotionCtrl, a unified and flexible motion controller for video generation designed to effectively and independently control camera and object motion. The architecture and training strategy of MotionCtrl are carefully devised, taking into account the inherent properties of camera motion, object motion, and imperfect training data. Compared to previous methods, MotionCtrl offers three main advantages: 1) It effectively and independently controls camera motion and object motion, enabling more fine-grained motion control and facilitating flexible and diverse combinations of both types of motion. 2) Its motion conditions are determined by camera poses and trajectories, which are appearance-free and minimally impact the appearance or shape of objects in generated videos. 3) It is a relatively generalizable model that can adapt to a wide array of camera poses and trajectories once trained. Extensive qualitative and quantitative experiments have been conducted to demonstrate the superiority of MotionCtrl over existing methods. Project Page: https://wzhouxiff.github.io/projects/MotionCtrl/

For agents

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

get_harvested_code_for_paper("2312.03641")
get_code_for_paper("2312.03641")
have("2312.03641")

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

curl https://syntology.ai/api/ran/2312.03641.json

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

[![Syntology run record](https://syntology.ai/api/ran/2312.03641.svg)](https://syntology.ai/paper/2312.03641)

Connect an agent — have() is free.