SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2305.20091 · ICCV · 2023

Humans in 4D: Reconstructing and Tracking Humans with Transformers

Jitendra Malik, Angjoo Kanazawa, Georgios Pavlakos, Jathushan Rajasegaran, Shubham Goel

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 (1 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 4 have no contract verdict; 0 failures of our instrument; 1 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/2305.20091.json · badge: /api/ran/2305.20091.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
shubham-goel/4D-Humans marked official in Papers with Code
Papers with Code archive
4 ran · 1 no recorded run
FunctionStatusWhere it lives
Discriminator Ran
ran on a synthesized input
shubham-goel/4D-Humans/hmr2/models/hmr2.py
code served (permissive licence) · get_code("044509f32558dfd9")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 2, 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 torch
import torch.nn as nn

# The Discriminator __init__ takes no arguments beyond self
INIT_ARGS = {}

# The forward method expects:
# - poses: (B, 23, 3, 3) - batch of SMPL body poses
# - betas: (B, 10) - batch of SMPL beta coefficients

batch_size = 2
num_joints = 23

poses = torch.randn(batch_size, num_joints, 3, 3)
betas = torch.randn(batch_size, 10)

ARGS = {
    'poses': poses,
    'betas': betas
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
Keypoint2DLoss Ran
ran on a synthesized input · constructed an object; no result was computed
shubham-goel/4D-Humans/hmr2/models/hmr2.py
code served (permissive licence) · get_code("a2086bce04d065c0")
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 an instance of its own class, Keypoint2DLoss: the run constructed the object and did not call it, so no result was computed. 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)

# Construct tensors for the forward pass
batch_size = 2
num_samples = 3
num_keypoints = 4

# pred_keypoints_2d: shape [B, S, N, 2]
pred_keypoints_2d = torch.randn(batch_size, num_samples, num_keypoints, 2)

# gt_keypoints_2d: shape [B, S, N, 3] (x, y, confidence)
gt_keypoints_2d = torch.randn(batch_size, num_samples, num_keypoints, 3)

# Ensure confidence values are in valid range [0, 1]
gt_keypoints_2d[:, :, :, 2] = torch.rand(batch_size, num_samples, num_keypoints)

ARGS = {
    'loss_type': 'l1'
}
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 Keypoint2DLoss — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Keypoint2DLoss 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
Keypoint3DLoss Ran
ran on a synthesized input
shubham-goel/4D-Humans/hmr2/models/hmr2.py
code served (permissive licence) · get_code("b0ad64515a64338e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 0, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

# INIT_ARGS for Keypoint3DLoss.__init__
INIT_ARGS = {
    'loss_type': 'l1'
}

# ARGS for Keypoint3DLoss.forward
# pred_keypoints_3d: [B, S, N, 3] -> [2, 5, 44, 3]
# gt_keypoints_3d: [B, S, N, 4] -> [2, 5, 44, 4] (last dim is confidence)
# pelvis_id: int (default 39, must be < N)

batch_size = 2
num_samples = 5
num_keypoints = 44
pelvis_id = 39

pred_keypoints_3d = torch.randn(batch_size, num_keypoints, 3)
gt_keypoints_3d = torch.randn(batch_size, num_keypoints, 4)
# Set confidence values to be between 0 and 1
gt_keypoints_3d[:, :, -1] = torch.rand(batch_size, num_keypoints)

ARGS = {
    'pred_keypoints_3d': pred_keypoints_3d,
    'gt_keypoints_3d': gt_keypoints_3d,
    'pelvis_id': pelvis_id
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 0, dtype family float — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
ParameterLoss Ran
ran on a synthesized input
shubham-goel/4D-Humans/hmr2/models/hmr2.py
code served (permissive licence) · get_code("8afa0da053f74850")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 0, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

# INIT_ARGS: ParameterLoss.__init__ takes no parameters
INIT_ARGS = {}

# ARGS: forward method parameters
# pred_param: [B, S, ...] - using [2, 3, 4] for batch=2, sequence=3, param_dim=4
# gt_param: same shape as pred_param
# has_param: [B] - binary mask indicating which samples have ground truth
batch_size = 2
seq_len = 3
param_dim = 4

pred_param = torch.randn(batch_size, seq_len, param_dim)
gt_param = torch.randn(batch_size, seq_len, param_dim)
has_param = torch.tensor([1.0, 0.0])  # first sample has GT, second doesn't

ARGS = {
    'pred_param': pred_param,
    'gt_param': gt_param,
    'has_param': has_param
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 0, dtype family float — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
HMR2 Not yet run
unverified (harvested, no recorded run)
shubham-goel/4D-Humans/hmr2/models/hmr2.py
code served (permissive licence) · get_code("c3db9cdd4369c570")
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, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 5 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.

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: A "transformerized" view of Human Mesh Recovery. We describe HMR 2.0, a fully transformer-based approach for 3D human pose and shape reconstruction from a single image. Besides impressive performance across a wide variety of poses and viewpoints, HMR 2.0 also acts as the backbone of an improved system for jointly reconstructing and tracking Humans in 4D (4DHumans). Here, we see output reconstructions from HMR 2.0 for each 2D detection in the left image.

For agents

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

get_harvested_code_for_paper("2305.20091")
get_code_for_paper("2305.20091")
have("2305.20091")

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

curl https://syntology.ai/api/ran/2305.20091.json

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

[![Syntology run record](https://syntology.ai/api/ran/2305.20091.svg)](https://syntology.ai/paper/2305.20091)

Connect an agent — have() is free.