SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2207.13784 · ECCV · 2022

AvatarPoser: Articulated Full-Body Pose Tracking from Sparse Motion Sensing

Christian Holz, Paul Streli, Jiaxi Jiang, Huajian Qiu, Andreas Fender, Larissa Laich, Patrick Snape

arXiv · PDF · Open in the Atlas

Code that ran

6 of 10 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; 4 failures of our instrument; 4 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/2207.13784.json · badge: /api/ran/2207.13784.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
eth-siplab/avatarposer canonical (classifier label)
classifier basis not recorded on the link
6 ran · 4 no recorded run
FunctionStatusWhere it lives
define_Model Ran
ran on a synthesized input
eth-siplab/avatarposer/models/select_model.py
code served (permissive licence) · get_code("f30ab21c7e7c1e81")
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 ModelAvatarPoser. 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)

# Create a minimal ModelAvatarPoser class for testing
class ModelAvatarPoser(nn.Module):
    def __init__(self, opt):
        super(ModelAvatarPoser, self).__init__()
        self.opt = opt
        # Add minimal layers to make it a valid nn.Module
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)

# Monkey-patch the import path
import sys
from types import ModuleType

models_module = ModuleType('models')
model_avatarposer_module = ModuleType('model_avatarposer')
model_avatarposer_module.ModelAvatarPoser = ModelAvatarPoser

models_module.model_avatarposer = model_avatarposer_module
sys.modules['models'] = models_module
sys.modules['models.model_avatarposer'] = model_avatarposer_module

# Construct the opt dict with required parameters
opt = {
    'model': 'avatarposer',
    'input_dim': 4,
    'output_dim': 3,
    'hidden_dim': 8,
}

ARGS = {
    'opt': opt
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
models, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ModelAvatarPoser — a ModelAvatarPoser output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ModelAvatarPoser 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
dict2str Ran
ran; our contract draft was wrong, not the code
eth-siplab/avatarposer/utils/utils_option.py
code served (permissive licence) · get_code("16c3e584beb5ce07")
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) and returned str. 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)

# Analyze the function:
# - dict2str takes 'opt' (a dict) and 'indent_l' (an int, default 1)
# - It recursively processes the dict, handling nested dicts and other values
# - It converts values to strings and formats them with indentation

# Create a test dictionary with nested structure
opt = {
    'learning_rate': 0.001,
    'batch_size': 32,
    'model': {
        'hidden_dim': 128,
        'num_layers': 3,
        'dropout': 0.5,
        'architecture': {
            'type': 'transformer',
            'heads': 8
        }
    },
    'epochs': 100,
    'optimizer': 'adam'
}

ARGS = {
    'opt': opt,
    'indent_l': 1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
find_last_checkpoint Ran
ran; our contract draft was wrong, not the code
eth-siplab/avatarposer/utils/utils_option.py
code served (permissive licence) · get_code("7f17c0641cd21670")
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, ndim 1, 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 os
import re
import glob
import tempfile
import shutil

# Create a temporary directory for testing
temp_dir = tempfile.mkdtemp()

# Create some dummy checkpoint files to test the function
checkpoint_files = [
    os.path.join(temp_dir, '100_G.pth'),
    os.path.join(temp_dir, '200_G.pth'),
    os.path.join(temp_dir, '50_G.pth'),
]

for checkpoint_file in checkpoint_files:
    with open(checkpoint_file, 'w') as f:
        f.write('dummy checkpoint')

ARGS = {
    'save_dir': temp_dir,
    'net_type': 'G',
}
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 tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
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_model_activation Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
eth-siplab/avatarposer/utils/utils_modelsummary.py
pointer only (licence: MIT); the stored copy get_code serves, from guanys-dar/mgdn, is pointer only (licence: NONE) · get_code("236d44054916d8c6")
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 guanys-dar/mgdn, 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.

gradient_penalty_loss Ran
ran; our contract draft was wrong, not the code
eth-siplab/avatarposer/models/loss.py
code served (permissive licence) · get_code("370bd3c6bf3bd496")
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 Tensor, 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 torch
import torch.nn as nn
from torch import autograd

torch.manual_seed(0)

class SimpleDiscriminator(nn.Module):
    """Minimal discriminator that takes 4D input and outputs scalar."""
    def __init__(self):
        super().__init__()
        self.fc = nn.Linear(4, 1)
    
    def forward(self, x):
        # Flatten spatial dimensions, keep batch
        batch_size = x.size(0)
        x_flat = x.view(batch_size, -1)
        return self.fc(x_flat)

discriminator = SimpleDiscriminator()

# Create batch_size=2, with spatial dimensions 1x1, feature dim=4
batch_size = 2
real_data = torch.randn(batch_size, 4, 1, 1)
fake_data = torch.randn(batch_size, 4, 1, 1)

# Optional weight tensor with same batch dimension
weight = torch.ones(batch_size, 1)

ARGS = {
    "discriminator": discriminator,
    "real_data": real_data,
    "fake_data": fake_data,
    "weight": weight,
}
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 Tensor, 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 Ran
ran on a synthesized input
eth-siplab/avatarposer/utils/utils_option.py
code served (permissive licence) · get_code("f15df57ab75e961b")
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 OrderedDict. 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 tempfile
import json
from collections import OrderedDict

# Create a temporary JSON configuration file
temp_dir = tempfile.mkdtemp()
opt_path = os.path.join(temp_dir, "options.json")

# Create a minimal valid configuration that matches what the function expects
config = {
    "task": "test_task",
    "scale": 4,
    "gpu_ids": [0],
    "datasets": {
        "train": {
            "name": "TrainDataset"
        },
        "val": {
            "name": "ValDataset"
        }
    },
    "path": {
        "root": temp_dir,
        "pretrained_netG": os.path.join(temp_dir, "pretrained.pth")
    },
    "netG": {
        "net_type": "RRDB",
        "in_nc": 3,
        "out_nc": 3,
        "nf": 64,
        "nb": 23
    },
    "train": {
        "pixel_opt": {
            "type": "l1"
        },
        "optim_g": {
            "type": "Adam",
            "lr": 0.0002
        }
    }
}

# Write config to file with comments that will be stripped
with open(opt_path, 'w') as f:
    for key, value in config.items():
        f.write(f'  "{key}": {json.dumps(value)},  // comment\n')
    # Remove trailing comma from last entry (fix JSON)
    
# Rewrite properly formatted JSON
with open(opt_path, 'w') as f:
    json.dump(config, f)

ARGS = {
    "opt_path": opt_path,
    "is_train": True
}
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 OrderedDict — a OrderedDict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a OrderedDict 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
g_path_regularize Not yet run
unverified (harvested, no recorded run)
eth-siplab/avatarposer/models/loss.py
code served (permissive licence) · get_code("d290bf91cc78c038")
How it ran

no invocation recorded.

get_model_complexity_info Not yet run
unverified (harvested, no recorded run)
eth-siplab/avatarposer/utils/utils_modelsummary.py
pointer only (licence: MIT); the stored copy get_code serves, from guanys-dar/mgdn, is pointer only (licence: NONE) · get_code("6ef91108ba92f025")
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 guanys-dar/mgdn, 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. No invocation recorded.

get_model_flops Not yet run
unverified (harvested, no recorded run)
eth-siplab/avatarposer/utils/utils_modelsummary.py
pointer only (licence: MIT); the stored copy get_code serves, from guanys-dar/mgdn, is pointer only (licence: NONE) · get_code("e39e724b057d5bec")
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 guanys-dar/mgdn, 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. No invocation recorded.

r1_penalty Not yet run
unverified (harvested, no recorded run)
eth-siplab/avatarposer/models/loss.py
code served (permissive licence) · get_code("7dfb6088a1ae2d27")
How it ran

no invocation recorded.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 8 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 0 the same month, 8 later) and 2 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
dict2str 23 other papers
6 dated earlier, 16 later, 1 same year, order not determinable
1905.02716 (2019-05)
EDVR: Video Restoration with Enhanced Deformable Convolutional Networks
1907.03222 (2019-07)
IRNet: A General Purpose Deep Residual Regression Framework for Materials Discovery
2005.05650 (2020-05)
Invertible Image Rescaling
and 20 more not named here
g_path_regularize 18 other papers
18 later
2303.17783 (2023-03)
Uncertainty-Aware Source-Free Adaptive Image Super-Resolution with Wavelet Augmentation Transformer
2305.07015 (2023-05)
Exploiting Diffusion Prior for Real-World Image Super-Resolution
2308.03448 (2023-08)
Make Explicit Calibration Implicit: Calibrate Denoiser Instead of the Noise Model
and 15 more not named here
gradient_penalty_loss 18 other papers
18 later
2303.17783 (2023-03)
Uncertainty-Aware Source-Free Adaptive Image Super-Resolution with Wavelet Augmentation Transformer
2305.07015 (2023-05)
Exploiting Diffusion Prior for Real-World Image Super-Resolution
2308.03448 (2023-08)
Make Explicit Calibration Implicit: Calibrate Denoiser Instead of the Noise Model
and 15 more not named here
r1_penalty 18 other papers
18 later
2303.17783 (2023-03)
Uncertainty-Aware Source-Free Adaptive Image Super-Resolution with Wavelet Augmentation Transformer
2305.07015 (2023-05)
Exploiting Diffusion Prior for Real-World Image Super-Resolution
2308.03448 (2023-08)
Make Explicit Calibration Implicit: Calibrate Denoiser Instead of the Noise Model
and 15 more not named here
find_last_checkpoint 5 other papers
2 dated earlier, 3 later
2103.14006 (2021-03)
Designing a Practical Degradation Model for Deep Blind Image Super-Resolution
2106.15368 (2021-06)
Text Prior Guided Scene Text Image Super-resolution
2308.08855 (2023-08)
Realistic Full-Body Tracking from Sparse Observations via Joint-Level Modeling
and 2 more not named here
get_model_activation 4 other papers
1 dated earlier, 3 later
2205.05675 (2022-05)
NTIRE 2022 Challenge on Efficient Super-Resolution: Methods and Results
2211.05910 (2022-11)
Efficient and Accurate Quantized Image Super-Resolution on Mobile NPUs, Mobile AI & AIM 2022 challenge: Report
2303.09472 (2023-03)
DiffIR: Efficient Diffusion Model for Image Restoration
and 1 more not named here
get_model_complexity_info 4 other papers
1 dated earlier, 3 later
2205.05675 (2022-05)
NTIRE 2022 Challenge on Efficient Super-Resolution: Methods and Results
2211.05910 (2022-11)
Efficient and Accurate Quantized Image Super-Resolution on Mobile NPUs, Mobile AI & AIM 2022 challenge: Report
2303.09472 (2023-03)
DiffIR: Efficient Diffusion Model for Image Restoration
and 1 more not named here
get_model_flops 4 other papers
2 dated earlier, 2 later
2201.06776 (2022-01)
Pruning-aware Sparse Regularization for Network Pruning
2205.05675 (2022-05)
NTIRE 2022 Challenge on Efficient Super-Resolution: Methods and Results
2303.09472 (2023-03)
DiffIR: Efficient Diffusion Model for Image Restoration
and 1 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

Today's Mixed Reality head-mounted displays track the user's head pose in world space as well as the user's hands for interaction in both Augmented Reality and Virtual Reality scenarios. While this is adequate to support user input, it unfortunately limits users' virtual representations to just their upper bodies. Current systems thus resort to floating avatars, whose limitation is particularly evident in collaborative settings. To estimate full-body poses from the sparse input sources, prior work has incorporated additional trackers and sensors at the pelvis or lower body, which increases setup complexity and limits practical application in mobile settings. In this paper, we present AvatarPoser, the first learning-based method that predicts full-body poses in world coordinates using only motion input from the user's head and hands. Our method builds on a Transformer encoder to extract deep features from the input signals and decouples global motion from the learned local joint orientations to guide pose estimation. To obtain accurate full-body motions that resemble motion capture animations, we refine the arm joints' positions using an optimization routine with inverse kinematics to match the original tracking input. In our evaluation, AvatarPoser achieved new state-of-the-art results in evaluations on large motion capture datasets (AMASS). At the same time, our method's inference speed supports realtime operation, providing a practical interface to support holistic avatar control and representation for Metaverse applications.

For agents

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

get_harvested_code_for_paper("2207.13784")
get_code_for_paper("2207.13784")
have("2207.13784")

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

curl https://syntology.ai/api/ran/2207.13784.json

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

[![Syntology run record](https://syntology.ai/api/ran/2207.13784.svg)](https://syntology.ai/paper/2207.13784)

Connect an agent — have() is free.