SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1802.09232 · 2018

2D/3D Pose Estimation and Action Recognition using Multitask Deep Learning

arXiv · PDF · Open in the Atlas

Code that ran

5 of 25 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 5 have no contract verdict; 0 failures of our instrument; 20 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/1802.09232.json · badge: /api/ran/1802.09232.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
dluvizon/deephar community (classifier label)
classifier basis not recorded on the link
5 ran · 20 no recorded run
FunctionStatusWhere it lives
channel_softmax_2d Ran
ran on a synthesized input
dluvizon/deephar/deephar/activations.py
code served (permissive licence) · get_code("3736dce1115c2cd2")
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. 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
import tensorflow as tf
from keras import backend as K

# Set seed for reproducibility
torch.manual_seed(0)
tf.random.set_seed(0)

# Create test tensors with ndim 4 and 5 as required by the function
# The function expects tensors where it can compute softmax over spatial dimensions (-3, -2)

# For ndim=4: (batch, channels, height, width)
test_tensor_4d = tf.constant(
    np.random.randn(2, 3, 4, 5).astype(np.float32)
)

# For ndim=5: (batch, channels, depth, height, width)
test_tensor_5d = tf.constant(
    np.random.randn(2, 3, 2, 4, 5).astype(np.float32)
)

# The function channel_softmax_2d only takes one parameter: alpha
# alpha is a scaling factor (default 1) applied to the input before softmax
# We'll test with alpha=1 (no scaling) as the simplest case

ARGS = {
    'alpha': 1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
keras, numpy, tensorflow, 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
mean_distance_error Ran
ran on a synthesized input
dluvizon/deephar/deephar/measures.py
code served (permissive licence) · get_code("6addcfc360155a19")
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 float64, 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 numpy as np

# Create small test data for mean_distance_error
# Function signature: mean_distance_error(y_true, y_pred)
# Both parameters need shape [num_samples, nb_joints, dim]

num_samples = 2
nb_joints = 3
dim = 4

# Create y_true with valid joint values (all > -1000000.0)
y_true = np.random.randn(num_samples, nb_joints, dim).astype(np.float32) * 10

# Create y_pred with similar shape
y_pred = np.random.randn(num_samples, nb_joints, dim).astype(np.float32) * 10

ARGS = {
    "y_true": y_true,
    "y_pred": y_pred,
}
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 float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
pck3d Ran
ran on a synthesized input
dluvizon/deephar/deephar/measures.py
code served (permissive licence) · get_code("365c7a3c29e72751")
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 float64, 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 numpy as np

# Create sample data for y_true and y_pred
# Shape: [num_samples, nb_joints, 3]
# num_samples=2, nb_joints=17, 3D coordinates
num_samples = 2
nb_joints = 17

# Create y_true with valid joint positions
y_true = np.random.randn(num_samples, nb_joints, 3) * 100 + 500
# Ensure all values are > -1000000 (the min_valid threshold)
y_true = np.abs(y_true) + 1.0

# Create y_pred with similar shape but slightly different values
y_pred = y_true + np.random.randn(num_samples, nb_joints, 3) * 10
# Ensure all values are > -1000000
y_pred = np.abs(y_pred) + 1.0

# Set refp parameter (threshold for PCK3D)
refp = 150

ARGS = {
    "y_true": y_true,
    "y_pred": y_pred,
    "refp": refp
}
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 float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
pckh Ran
ran on a synthesized input
dluvizon/deephar/deephar/measures.py
code served (permissive licence) · get_code("14edab500394ebe9")
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 float64, 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 numpy as np

# Set seed for reproducibility
np.random.seed(0)

# Create test data
num_samples = 2
nb_joints = 16
num_used_joints = 14

# y_true: [num_samples, nb_joints, 2]
y_true = np.random.rand(num_samples, nb_joints, 2) * 100

# y_pred: [num_samples, nb_joints, 2]
y_pred = y_true + np.random.rand(num_samples, nb_joints, 2) * 5

# head_size: [num_samples, 1]
head_size = np.array([[20.0], [25.0]])

ARGS = {
    "y_true": y_true,
    "y_pred": y_pred,
    "head_size": head_size,
    "refp": 0.5
}
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 float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
relu Ran
ran on a synthesized input
dluvizon/deephar/deephar/layers.py
code served (permissive licence) · get_code("401ff7831c3db2d1")
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 KerasTensor. 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 keras
import numpy as np

# Create a small input tensor for the function
# The function expects a Keras tensor or similar input
input_data = np.random.randn(2, 4).astype(np.float32)
x = keras.Input(shape=(4,))

ARGS = {
    'x': x,
    'leakyrelu': False,
    'name': 'relu_activation'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
keras, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type KerasTensor — a KerasTensor 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
action_prediction_early_fusion Not yet run
unverified (harvested, no recorded run)
dluvizon/deephar/deephar/models/spnet.py
code served (permissive licence) · get_code("2b23b3aa62c445cb")
How it ran

no invocation recorded.

action_top Not yet run
unverified (harvested, no recorded run)
dluvizon/deephar/deephar/models/action.py
code served (permissive licence) · get_code("d103639c80df1e4f")
How it ran

no invocation recorded.

add_tensorlist Not yet run
unverified (harvested, no recorded run)
dluvizon/deephar/deephar/models/common.py
code served (permissive licence) · get_code("0cf7210017234f9c")
How it ran

no invocation recorded.

build_act_pred_block Not yet run
unverified (harvested, no recorded run)
dluvizon/deephar/deephar/models/action.py
code served (permissive licence) · get_code("fe05101a124ed6e5")
How it ran

no invocation recorded.

build_pose_model Not yet run
unverified (harvested, no recorded run)
dluvizon/deephar/deephar/models/action.py
code served (permissive licence) · get_code("a7634fec4f5ad0a3")
How it ran

no invocation recorded.

build_reception_block Not yet run
unverified (harvested, no recorded run)
dluvizon/deephar/deephar/models/reception.py
code served (permissive licence) · get_code("6b2a3d9fe907a974")
How it ran

no invocation recorded.

build_regmap_block Not yet run
unverified (harvested, no recorded run)
dluvizon/deephar/deephar/models/reception.py
code served (permissive licence) · get_code("63e1b7fec39396c5")
How it ran

no invocation recorded.

build_sconv_block Not yet run
unverified (harvested, no recorded run)
dluvizon/deephar/deephar/models/reception.py
code served (permissive licence) · get_code("d50b506eab82722e")
How it ran

no invocation recorded.

concat_tensorlist Not yet run
unverified (harvested, no recorded run)
dluvizon/deephar/deephar/models/common.py
code served (permissive licence) · get_code("bdfe57e9d50bfd4f")
How it ran

no invocation recorded.

conv2d Not yet run
unverified (harvested, no recorded run)
dluvizon/deephar/deephar/layers.py
code served (permissive licence) · get_code("7bbc7651f4df590a")
How it ran

no invocation recorded.

conv_block Not yet run
unverified (harvested, no recorded run)
dluvizon/deephar/deephar/models/blocks.py
code served (permissive licence) · get_code("dd1d092a82873639")
How it ran

no invocation recorded.

elasticnet_bincross_loss_on_valid_joints Not yet run
unverified (harvested, no recorded run)
dluvizon/deephar/deephar/losses.py
code served (permissive licence) · get_code("33bbcc52a7b5557f")
How it ran

no invocation recorded.

elasticnet_loss_on_valid_joints Not yet run
unverified (harvested, no recorded run)
dluvizon/deephar/deephar/losses.py
code served (permissive licence) · get_code("187971aca9ed7375")
How it ran

no invocation recorded.

elasticnet_loss_on_valid_joints Not yet run
unverified (harvested, no recorded run)
dluvizon/deephar/deephar/objectives.py
code served (permissive licence) · get_code("e828ec5077b00a62")
How it ran

no invocation recorded.

identity_block Not yet run
unverified (harvested, no recorded run)
dluvizon/deephar/deephar/models/blocks.py
code served (permissive licence) · get_code("4310717bca3c3fee")
How it ran

no invocation recorded.

l1_loss_on_valid_joints Not yet run
unverified (harvested, no recorded run)
dluvizon/deephar/deephar/losses.py
code served (permissive licence) · get_code("5cd044b993eb22d7")
How it ran

no invocation recorded.

prediction_block Not yet run
unverified (harvested, no recorded run)
dluvizon/deephar/deephar/models/spnet.py
code served (permissive licence) · get_code("fc7597186a5bf1a3")
How it ran

no invocation recorded.

prediction_branch Not yet run
unverified (harvested, no recorded run)
dluvizon/deephar/deephar/models/spnet.py
code served (permissive licence) · get_code("da27fcfda5118da4")
How it ran

no invocation recorded.

residual_unit Not yet run
unverified (harvested, no recorded run)
dluvizon/deephar/deephar/models/common.py
code served (permissive licence) · get_code("56b0283b329ccb24")
How it ran

no invocation recorded.

stem_inception_v4 Not yet run
unverified (harvested, no recorded run)
dluvizon/deephar/deephar/models/blocks.py
code served (permissive licence) · get_code("0861e93b7e6d2418")
How it ran

no invocation recorded.

Also present in other papers' code

Of 25 samples lifted from this paper's linked repositories, 25 carry a function identity and 0 do not. Of the 25 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 25 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 2026-11-21.

Repositories linked to this paper

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

Abstract

Action recognition and human pose estimation are closely related but both problems are generally handled as distinct tasks in the literature. In this work, we propose a multitask framework for jointly 2D and 3D pose estimation from still images and human action recognition from video sequences. We show that a single architecture can be used to solve the two problems in an efficient way and still achieves state-of-the-art results. Additionally, we demonstrate that optimization from end-to-end leads to significantly higher accuracy than separated learning. The proposed architecture can be trained with data from different categories simultaneously in a seamlessly way. The reported results on four datasets (MPII, Human3.6M, Penn Action and NTU) demonstrate the effectiveness of our method on the targeted tasks.

For agents

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

get_harvested_code_for_paper("1802.09232")
get_code_for_paper("1802.09232")
have("1802.09232")

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

curl https://syntology.ai/api/ran/1802.09232.json

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

[![Syntology run record](https://syntology.ai/api/ran/1802.09232.svg)](https://syntology.ai/paper/1802.09232)

Connect an agent — have() is free.