SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2209.09484 · 2022

Hierarchical Temporal Transformer for 3D Hand Pose Estimation and Action Recognition from Egocentric RGB Videos

arXiv · PDF · Open in the Atlas

Code that ran

7 of 8 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 4 have no contract verdict; 3 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/2209.09484.json · badge: /api/ran/2209.09484.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
fylwen/htt canonical (classifier label)
classifier basis not recorded on the link
7 ran · 1 no recorded run
FunctionStatusWhere it lives
conv3x3 Ran
ran; our contract draft was wrong, not the code
fylwen/htt/models/resnet.py
pointer only (licence: MIT); 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 (MIT), 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.

load_opts Ran
ran on a synthesized input
fylwen/htt/netscripts/reloadmodel.py
code served (permissive licence) · get_code("d84e7bf13e0f97c7")
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 shutil

# Create a temporary directory to simulate the checkpoint structure
temp_dir = tempfile.mkdtemp()

# Create a subdirectory to act as the checkpoint directory
checkpoint_dir = os.path.join(temp_dir, 'checkpoints')
os.makedirs(checkpoint_dir, exist_ok=True)

# Create a sample opts object (can be any picklable object)
sample_opts = {
    'learning_rate': 0.001,
    'batch_size': 32,
    'epochs': 100,
    'model_name': 'test_model'
}

# Save the opts to opt.pkl in the checkpoint directory
opt_pkl_path = os.path.join(checkpoint_dir, 'opt.pkl')
with open(opt_pkl_path, 'wb') as f:
    pickle.dump(sample_opts, f)

# Create the resume_checkpoint path (absolute path format)
resume_checkpoint = checkpoint_dir

ARGS = {
    'resume_checkpoint': resume_checkpoint
}
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
parse_evaluators Ran
ran on a synthesized input
fylwen/htt/netscripts/position_evaluator.py
code served (permissive licence) · get_code("d921e2ab2e6c7662")
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

torch.manual_seed(0)

class MockEvaluator:
    """Mock evaluator with get_measures method"""
    def get_measures(self, start, end, steps):
        # Return dummy values matching expected output format
        epe_mean = np.array(0.5)
        epe_mean_joints = np.array([0.4, 0.6, 0.5, 0.7])
        epe_median = np.array(0.45)
        auc = np.array(0.8)
        thresholds = np.linspace(start, end, steps)
        pck_curve = np.linspace(0.0, 1.0, steps)
        return epe_mean, epe_mean_joints, epe_median, auc, pck_curve, thresholds

evaluators = {
    'joints3d': MockEvaluator(),
    'left_joints3d': MockEvaluator(),
    'right_joints3d': MockEvaluator(),
    'left_joints3d_cent': MockEvaluator(),
    'right_joints3d_cent': MockEvaluator(),
    'joints3d_cent': MockEvaluator(),
}

config = {
    'joints3d': [0, 0.08, 33],
    'left_joints3d': [0, 0.09, 37],
    'right_joints3d': [0, 0.09, 37],
    'left_joints3d_cent': [0, 0.05, 21],
    'right_joints3d_cent': [0, 0.05, 21],
    'joints3d_cent': [0, 0.05, 21]
}

ARGS = {
    'evaluators': evaluators,
    'config': config,
}
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
recover_3d_proj_pinhole Ran
ran on a synthesized input
fylwen/htt/models/utils.py
code served (permissive licence) · get_code("327766cdbd033189")
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

torch.manual_seed(0)

# Construct arguments for recover_3d_proj_pinhole
batch_size = 2
num_joints = 4

# camintr: camera intrinsic matrix [batch_size, 3, 3]
# The function accesses camintr[:, :1, :1] for focal length
# and camintr[:, :2, 2] for camera center
camintr = torch.tensor([
    [[500.0, 0.0, 64.0],
     [0.0, 500.0, 64.0],
     [0.0, 0.0, 1.0]],
    [[520.0, 0.0, 64.0],
     [0.0, 520.0, 64.0],
     [0.0, 0.0, 1.0]]
], dtype=torch.float32)

# est_scale: estimated scale [batch_size, num_joints]
est_scale = torch.randn(batch_size, num_joints, dtype=torch.float32)

# est_trans: estimated translation [batch_size, num_joints, 2]
est_trans = torch.randn(batch_size, num_joints, 2, dtype=torch.float32)

# off_z: offset for Z (default parameter)
off_z = 0.4

# input_res: input resolution tuple
input_res = (128, 128)

# verbose: boolean flag
verbose = False

ARGS = {
    'camintr': camintr,
    'est_scale': est_scale,
    'est_trans': est_trans,
    'off_z': off_z,
    'input_res': input_res,
    'verbose': verbose
}
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
resnet18 Ran
ran; our contract draft was wrong, not the code
fylwen/htt/models/resnet.py
code served (permissive licence) · get_code("75131af8f1e8c065")
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 ResNet. 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)

ARGS = {
    'pretrained': False
}
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 ResNet — a ResNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet 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
resnet34 Ran
ran; our contract draft was wrong, not the code
fylwen/htt/models/resnet.py
code served (permissive licence) · get_code("abce9f8e2e1ff31d")
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 ResNet. 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)

ARGS = {
    'pretrained': False
}
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 ResNet — a ResNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet 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
torch2numpy Ran
ran on a synthesized input
fylwen/htt/models/utils.py
code served (permissive licence) · get_code("c543ebbcba3e5933")
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, 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

torch.manual_seed(0)

# The function torch2numpy takes an input parameter
# It checks if input is None, if it's a torch tensor, and converts it accordingly
# We need to provide test cases for each path

# Test case 1: A torch tensor on CPU
tensor_input = torch.randn(2, 4)

# Test case 2: None
none_input = None

# Test case 3: A non-tensor object (like a numpy array or regular Python object)
import numpy as np
numpy_input = np.array([[1, 2, 3, 4], [5, 6, 7, 8]])

# Since the function signature only has one parameter 'input',
# we'll test with a tensor (the most common case)
ARGS = {
    "input": tensor_input
}
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 ndarray, 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
compute_hand_loss Not yet run
unverified (harvested, no recorded run)
fylwen/htt/models/utils.py
code served (permissive licence) · get_code("6ba5634d0eca83b8")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 1 the same month, 3 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.

FunctionAlso present inEarliest-datedOthers named
conv3x3 686 other papers
411 dated earlier, 3 the same month, 261 later, 11 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
resnet18 89 other papers
53 dated earlier, 34 later, 2 same year, order not determinable
1705.09759 (2017-05)
CASENet: Deep Category-Aware Semantic Edge Detection
1801.07698 (2018-01)
ArcFace: Additive Angular Margin Loss for Deep Face Recognition
1803.00839 (2018-03)
Pose-Robust Face Recognition via Deep Residual Equivariant Mapping
and 86 more not named here
resnet34 53 other papers
39 dated earlier, 12 later, 2 same year, order not determinable
1705.09759 (2017-05)
CASENet: Deep Category-Aware Semantic Edge Detection
1801.07698 (2018-01)
ArcFace: Additive Angular Margin Loss for Deep Face Recognition
1803.00839 (2018-03)
Pose-Robust Face Recognition via Deep Residual Equivariant Mapping
and 50 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

Understanding dynamic hand motions and actions from egocentric RGB videos is a fundamental yet challenging task due to self-occlusion and ambiguity. To address occlusion and ambiguity, we develop a transformer-based framework to exploit temporal information for robust estimation. Noticing the different temporal granularity of and the semantic correlation between hand pose estimation and action recognition, we build a network hierarchy with two cascaded transformer encoders, where the first one exploits the short-term temporal cue for hand pose estimation, and the latter aggregates per-frame pose and object information over a longer time span to recognize the action. Our approach achieves competitive results on two first-person hand action benchmarks, namely FPHA and H2O. Extensive ablation studies verify our design choices.

For agents

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

get_harvested_code_for_paper("2209.09484")
get_code_for_paper("2209.09484")
have("2209.09484")

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

curl https://syntology.ai/api/ran/2209.09484.json

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

[![Syntology run record](https://syntology.ai/api/ran/2209.09484.svg)](https://syntology.ai/paper/2209.09484)

Connect an agent — have() is free.