SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2203.08063 · ECCV · 2022

MotionCLIP: Exposing Human Motion Generation to CLIP Space

Amit Bermano, Brian Gordon, Daniel Cohen-Or, Amir Hertz, Guy Tevet

arXiv · PDF · Open in the Atlas

Code that ran

4 of 5 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; 0 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 2 failures of our instrument; 1 no recorded run), measured 2026-11-21; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2203.08063.json · badge: /api/ran/2203.08063.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
guytevet/motionclip canonical (classifier label)
classifier basis not recorded on the link
4 ran · 1 no recorded run
FunctionStatusWhere it lives
get_datasets Ran
ran on a synthesized input
guytevet/motionclip/src/datasets/get_dataset.py
code served (permissive licence) · get_code("4a0130412b541041")
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 torch
import numpy as np
from unittest.mock import Mock

torch.manual_seed(0)

class MockPreprocess:
    """Stand-in for clip_preprocess callable"""
    def __call__(self, x):
        return x

class MockAMASS:
    """Stand-in for AMASS dataset class"""
    def __init__(self, split='train', clip_preprocess=None, **parameters):
        self.split = split
        self.clip_preprocess = clip_preprocess
        self.parameters = parameters
        self.data = torch.randn(2, 4)  # small mock data
    
    def update_parameters(self, parameters):
        self.parameters = parameters

# Monkey-patch AMASS into the function's namespace
import sys
current_module = sys.modules[__name__]
setattr(current_module, 'AMASS', MockAMASS)

# Build parameters dict
parameters = {
    'batch_size': 32,
    'num_frames': 60,
    'data_dir': '/tmp/mock_data',
    'normalize': True
}

# Build clip_preprocess callable
clip_preprocess = MockPreprocess()

ARGS = {
    'parameters': parameters,
    'clip_preprocess': clip_preprocess,
    'split': 'train'
}
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 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_trans_from_vibe Ran
ran on a synthesized input
guytevet/motionclip/src/datasets/amass.py
code served (permissive licence) · get_code("137a935fa5d0b871")
How it ran

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

import numpy as np

# Construct a minimal vibe dictionary that satisfies the function's requirements
# The function accesses:
# - vibe['joints3d']: shape (T, N, 2) where T is time frames, N is joints
# - vibe['orig_cam']: shape (T, 4) with columns [scale, ?, x, y]

num_frames = 2
num_joints = 5

# Create joints3d with shape (num_frames, num_joints, 2)
joints3d = np.random.randn(num_frames, num_joints, 2) * 100 + 50

# Create orig_cam with shape (num_frames, 4)
# Format: [cam_scale, unknown, cam_x, cam_y]
orig_cam = np.array([
    [1.0, 0.5, 100.0, 120.0],
    [1.1, 0.6, 105.0, 125.0]
])

vibe = {
    'joints3d': joints3d,
    'orig_cam': orig_cam
}

ARGS = {
    'vibe': vibe,
    'use_z': True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (2, 3) — shape (2, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_z Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
guytevet/motionclip/src/datasets/amass.py
code served (permissive licence) · get_code("e182600b67dc0a48")
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

# Construct valid arguments for get_z function
# The function expects:
# - cam_s: scaling factor (scalar or array-like)
# - cam_pos: camera position (2D array/list)
# - joints: joint positions (Nx2 or Nx3 array)
# - img_size: image size (scalar or array-like)
# - flength: focal length (scalar)

# Create small test data
np.random.seed(0)

# cam_s: camera scale factor (scalar)
cam_s = 1.5

# cam_pos: camera position in 2D (2-element array)
cam_pos = np.array([0.1, 0.2])

# joints: joint positions (N x 2 or N x 3 array)
# Using 5 joints with x, y, z coordinates
joints = np.array([
    [0.5, 0.3, 1.0],
    [0.6, 0.4, 1.1],
    [0.7, 0.5, 1.2],
    [0.55, 0.35, 1.05],
    [0.65, 0.45, 1.15]
], dtype=np.float32)

# img_size: image dimensions (scalar or 2-element array)
img_size = 224

# flength: focal length (scalar)
flength = 500.0

ARGS = {
    "cam_s": cam_s,
    "cam_pos": cam_pos,
    "joints": joints,
    "img_size": img_size,
    "flength": flength
}
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
parse_info_name Ran
ran; our contract draft was wrong, not the code
guytevet/motionclip/src/datasets/tools.py
code served (permissive licence) · get_code("7aee01d2ce301af5")
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 string
import tempfile

# Create a temporary file with a structured name for testing
temp_dir = tempfile.mkdtemp()
test_path = os.path.join(temp_dir, "a1b2c3.txt")

# Create the file so the path is valid
with open(test_path, 'w') as f:
    f.write("")

ARGS = {
    "path": test_path
}
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 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_dataset Not yet run
unverified (harvested, no recorded run)
guytevet/motionclip/src/datasets/get_dataset.py
code served (permissive licence) · get_code("94fefe5771a39978")
How it ran

no invocation recorded.

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 0 later) and 3 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

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

FunctionAlso present inEarliest-datedOthers named
get_z 1 other paper
1 dated earlier
2104.05670 (2021-04)
Action-Conditioned 3D Human Motion Synthesis with Transformer VAE
parse_info_name 1 other paper
1 dated earlier
2104.05670 (2021-04)
Action-Conditioned 3D Human Motion Synthesis with Transformer VAE

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-21.

Repositories linked to this paper

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

Abstract

We introduce MotionCLIP, a 3D human motion auto-encoder featuring a latent embedding that is disentangled, well behaved, and supports highly semantic textual descriptions. MotionCLIP gains its unique power by aligning its latent space with that of the Contrastive Language-Image Pre-training (CLIP) model. Aligning the human motion manifold to CLIP space implicitly infuses the extremely rich semantic knowledge of CLIP into the manifold. In particular, it helps continuity by placing semantically similar motions close to one another, and disentanglement, which is inherited from the CLIP-space structure. MotionCLIP comprises a transformer-based motion auto-encoder, trained to reconstruct motion while being aligned to its text label's position in CLIP-space. We further leverage CLIP's unique visual understanding and inject an even stronger signal through aligning motion to rendered frames in a self-supervised manner. We show that although CLIP has never seen the motion domain, MotionCLIP offers unprecedented text-to-motion abilities, allowing out-of-domain actions, disentangled editing, and abstract language specification. For example, the text prompt "couch" is decoded into a sitting down motion, due to lingual similarity, and the prompt "Spiderman" results in a web-swinging-like solution that is far from seen during training. In addition, we show how the introduced latent space can be leveraged for motion interpolation, editing and recognition. 1

For agents

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

get_harvested_code_for_paper("2203.08063")
get_code_for_paper("2203.08063")
have("2203.08063")

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

curl https://syntology.ai/api/ran/2203.08063.json

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

[![Syntology run record](https://syntology.ai/api/ran/2203.08063.svg)](https://syntology.ai/paper/2203.08063)

Connect an agent — have() is free.