SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1812.01601 · 2018

Learning 3D Human Dynamics from Video

arXiv · PDF · Open in the Atlas

Code that ran

2 of 7 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; 0 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/1812.01601.json · badge: /api/ran/1812.01601.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
akanazawa/human_dynamics canonical (classifier label)
classifier basis: census
2 ran · 5 no recorded run
FunctionStatusWhere it lives
float_feature Ran
ran on a synthesized input
akanazawa/human_dynamics/src/datasets/common.py
code served (permissive licence) · get_code("fed8c87d61cd6a4b")
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 Feature. 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 tensorflow as tf

# Test with a single float value
ARGS = {
    "value": 3.14
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Feature — a Feature output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Feature 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_all_data Ran
ran on a synthesized input
akanazawa/human_dynamics/src/datasets/h36_to_tfrecords_video.py
code served (permissive licence) · get_code("0575ea0e8e017a77")
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 tempfile
import pickle
import numpy as np
import os
from pathlib import Path

# Create a temporary directory structure
tmpdir = tempfile.mkdtemp()
img_dir = tmpdir

# Create dummy PNG files
for i in range(3):
    png_path = os.path.join(img_dir, f'image_{i}.png')
    Path(png_path).touch()

# Create gt_poses.pkl with mock 2D and 3D poses
# COMMON_JOINT_IDS has 14 joints
num_images = 3
num_all_joints = 17  # Typical pose format before filtering

# Create mock 2D poses (num_images x num_all_joints x 2)
gt2ds_mock = [np.random.rand(num_all_joints, 2) * 100 for _ in range(num_images)]

# Create mock 3D poses (num_images x num_all_joints x 3) in mm
gt3ds_mock = [np.random.rand(num_all_joints, 3) * 1000 for _ in range(num_images)]

gts = {
    '2d': gt2ds_mock,
    '3d': gt3ds_mock
}

gt_path = os.path.join(img_dir, 'gt_poses.pkl')
with open(gt_path, 'wb') as f:
    pickle.dump(gts, f)

# Create camera_wext.pkl with mock camera parameters
cam_mock = {
    'f': np.array([500.0, 500.0]),  # focal length in x and y
    'c': np.array([320.0, 240.0])   # principal point
}

cam_path = os.path.join(img_dir, 'camera_wext.pkl')
with open(cam_path, 'wb') as f:
    pickle.dump(cam_mock, f)

ARGS = {
    'img_dir': img_dir,
    'split': 'train'
}
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
compute_loss_e_3d Not yet run
unverified (harvested, no recorded run)
akanazawa/human_dynamics/src/ops.py
code served (permissive licence) · get_code("5bf26791a13b1b20")
How it ran

no invocation recorded.

compute_loss_e_kp Not yet run
unverified (harvested, no recorded run)
akanazawa/human_dynamics/src/ops.py
code served (permissive licence) · get_code("5adb3a6418860a82")
How it ran

no invocation recorded.

int64_feature Not yet run
unverified (harvested, no recorded run)
akanazawa/human_dynamics/src/datasets/common.py
code served (permissive licence) · get_code("e64a6b3b07db253d")
How it ran

no invocation recorded.

read_from_example Not yet run
unverified (harvested, no recorded run)
akanazawa/human_dynamics/src/datasets/common.py
code served (permissive licence) · get_code("e8e628d7220b85a2")
How it ran

no invocation recorded.

run_poseflow Not yet run
unverified (harvested, no recorded run)
akanazawa/human_dynamics/extract_tracks.py
code served (permissive licence) · get_code("d5f1c58706737981")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 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
read_from_example 1 other paper
1 later
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

From an image of a person in action, we can easily guess the 3D motion of the person in the immediate past and future. This is because we have a mental model of 3D human dynamics that we have acquired from observing visual sequences of humans in motion. We present a framework that can similarly learn a representation of 3D dynamics of humans from video via a simple but effective temporal encoding of image features. At test time, from video, the learned temporal representation give rise to smooth 3D mesh predictions. From a single image, our model can recover the current 3D mesh as well as its 3D past and future motion. Our approach is designed so it can learn from videos with 2D pose annotations in a semi-supervised manner. Though annotated data is always limited, there are millions of videos uploaded daily on the Internet. In this work, we harvest this Internet-scale source of unlabeled data by training our model on unlabeled video with pseudo-ground truth 2D pose obtained from an off-the-shelf 2D pose detector. Our experiments show that adding more videos with pseudo-ground truth 2D pose monotonically improves 3D prediction performance. We evaluate our model, Human Mesh and Motion Recovery (HMMR), on the recent challenging dataset of 3D Poses in the Wild and obtain state-of-the-art performance on the 3D prediction task without any fine-tuning. The project website with video, code, and data can be found at https://akanazawa.github.io/human_dynamics/.

For agents

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

get_harvested_code_for_paper("1812.01601")
get_code_for_paper("1812.01601")
have("1812.01601")

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

curl https://syntology.ai/api/ran/1812.01601.json

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

[![Syntology run record](https://syntology.ai/api/ran/1812.01601.svg)](https://syntology.ai/paper/1812.01601)

Connect an agent — have() is free.