Papers › Cross-Attention of Disentangled Modalities for 3D Human Mesh Recovery with Transformers

Cross-Attention of Disentangled Modalities for 3D Human Mesh Recovery with Transformers

27 Jul 2022arXiv:2207.13820archive 2025-07-28

Junhyeong Cho, Kim Youwang, Tae-Hyun Oh

Transformer encoder architectures have recently achieved state-of-the-art results on monocular 3D human mesh reconstruction, but they require a substantial number of parameters and expensive computations. Due to the large memory overhead and slow inference speed, it is difficult to deploy such models for practical use. In this paper, we propose a novel transformer encoder-decoder architecture for 3D human mesh reconstruction from a single image, called FastMETRO. We identify the performance bottleneck in the encoder-based transformers is caused by the token design which introduces high complexity interactions among input tokens. We disentangle the interactions via an encoder-decoder architecture, which allows our model to demand much fewer parameters and shorter inference time. In addition, we impose the prior knowledge of human body's morphological relationship via attention masking and mesh upsampling operations, which leads to faster convergence with higher accuracy. Our FastMETRO improves the Pareto-front of accuracy and efficiency, and clearly outperforms image-based methods on Human3.6M and 3DPW. Furthermore, we validate its generalizability on FreiHAND.

PaperPDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 16 code samples harvested from 1 repository linked to this paper: 12 ran (of which 4 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 4 where Syntology's instrument failed) · 4 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; 1 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 16 samples from 1 repository, 12 ran (8 with no instrument failure; 4 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

postech-ami/fastmetro officialmentioned in papermentioned on GitHubpytorchSyntology: 16 samples, 12 ran (4 instrument-failed) · 4 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

16 samples harvested: 12 ran (of which 4 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 4 where Syntology's instrument failed) · 4 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

3ran · our draft was wrong
1ran; the contract check could not run
8ran
4unverified

Licence: 0 of the 16 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from postech-ami/fastmetro. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. For 2 samples that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

postech-ami/fastmetro official repository 16 samples, 12 ran (4 instrument-failed) · 4 unverified
FCBlock in 8 papers src/modeling/model/modeling_fastmetro.py ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · 562c62242b4e4a11 · report fixture: returned FCBlock · level recorded 2026-09-22
FCResBlock src/modeling/model/modeling_fastmetro.py ran · metamorphic tier: deterministic MIT (permissive) · 4d8d11aa70b28920 · report fixture: returned FCResBlock · level recorded 2026-09-22
PositionEmbeddingSine in 21 papers
census: 6 implementations from 6 papers, 3 distinct outputs across 3 buckets, one shared input per bucket
src/modeling/model/modeling_fastmetro.py ran · metamorphic tier: deterministic MIT (permissive) · ab7c9f19268bf5b1 · report fixture: returned PositionEmbeddingSine · level recorded 2026-09-22
TransformerDecoder in 49 papers src/modeling/model/modeling_fastmetro.py ran MIT (permissive) · 4a7551bd532ddd12 · report fixture: signature tgt:Tensor/3/float, memory:Tensor/3/float, tgt_mask:NoneType/0/?, memory_mask:NoneType/0/?, tgt_key_padding_mask:NoneType/0/?, memory_key_padding_mask:NoneType/0/?, pos:NoneType/0/?, query_pos:NoneType/0/? · returned Tensor · level recorded 2026-09-24
TransformerDecoderLayer in 45 papers src/modeling/model/modeling_fastmetro.py ran MIT (permissive) · b7f02dc052c23ed5 · report fixture: signature tgt:Tensor/3/float, memory:Tensor/3/float, tgt_mask:NoneType/0/?, memory_mask:NoneType/0/?, tgt_key_padding_mask:NoneType/0/?, memory_key_padding_mask:NoneType/0/?, pos:Tensor/3/float, query_pos:Tensor/3/float · returned Tensor · level recorded 2026-09-24
TransformerEncoder in 96 papers src/modeling/model/modeling_fastmetro.py ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · f432265aed9e2e0b · report fixture: returned TransformerEncoder · level recorded 2026-09-22
TransformerEncoderLayer in 90 papers src/modeling/model/modeling_fastmetro.py ran MIT (permissive) · 81cad16cd0d5efb1 · report fixture: signature src:Tensor/3/float, src_mask:NoneType/0/?, src_key_padding_mask:Tensor/2/bool, pos:Tensor/3/float · returned Tensor · level recorded 2026-09-24
_get_activation_fn in 21 papers src/modeling/model/modeling_fastmetro.py ran · our draft was wrong MIT (permissive) · d493e14665ada3f2 · report fixture: signature activation:str/0/str · returned function · level recorded 2026-09-22
_get_clones in 28 papers src/modeling/model/modeling_fastmetro.py ran · our draft was wrong MIT (permissive) · 873e680b6cfcd7a5 · report fixture: signature module:callable/0/?, N:int/0/int · returned ModuleList · level recorded 2026-09-22
batch_svd src/modeling/model/modeling_fastmetro.py ran; the contract check could not run (RAISES) · metamorphic tier: deterministic fingerprinted MIT (permissive) · 6697b1ba4e63ee2a · report fixture: signature A:Tensor/3/float · returned tuple · level recorded 2026-09-22
build_position_encoding in 42 papers src/modeling/model/modeling_fastmetro.py ran · our draft was wrong MIT (permissive) · 4407bd06305a2f6c · report fixture: signature pos_type:str/0/str, hidden_dim:int/0/int · returned PositionEmbeddingSine · level recorded 2026-09-22
build_smpl_parameter_regressor src/modeling/model/modeling_fastmetro.py ran · metamorphic tier: deterministic MIT (permissive) · 52fdd40a81f5afe0 · report fixture: returned SMPL_Parameter_Regressor · level recorded 2026-09-22
FastMETRO_Body_Network src/modeling/model/modeling_fastmetro.py unverified MIT (permissive) · f8c49af84f39dc14 · report
SMPL_Parameter_Regressor src/modeling/model/modeling_fastmetro.py unverified MIT (permissive) · 12981f7af75b69b4 · report
Transformer in 173 papers src/modeling/model/modeling_fastmetro.py unverified MIT (permissive) · bc515da1433c761e · report
build_transformer in 62 papers src/modeling/model/modeling_fastmetro.py unverified MIT (permissive) · c6caa9a5fb3f06a7 · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for FastMETRO (graph method: FastMETRO). 16 bricks; individually on a fixture: 12 ran (of which 4 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 4 where Syntology's instrument failed) · 4 unverified. The assembly has not been run end-to-end.

  1. FastMETRO_Body_Network entry · level 0 · unverified · f8c49af84f39dc14
  2. build_position_encoding helper · level 1 · ran · our draft was wrong · 4407bd06305a2f6c
  3. build_smpl_parameter_regressor helper · level 1 · ran · 52fdd40a81f5afe0
  4. build_transformer helper · level 0 · unverified · c6caa9a5fb3f06a7
  5. PositionEmbeddingSine model · level 1 · ran · ab7c9f19268bf5b1
  6. SMPL_Parameter_Regressor model · level 0 · unverified · 12981f7af75b69b4
  7. Transformer model · level 0 · unverified · bc515da1433c761e
  8. FCBlock helper · level 1 · ran · 562c62242b4e4a11
  9. FCResBlock helper · level 1 · ran · 4d8d11aa70b28920
  10. batch_svd helper · level 1 · ran; the contract check could not run (RAISES) · 6697b1ba4e63ee2a
  11. TransformerDecoder model · level 1 · ran · 4a7551bd532ddd12
  12. TransformerDecoderLayer model · level 1 · ran · b7f02dc052c23ed5
  13. TransformerEncoder model · level 1 · ran · f432265aed9e2e0b
  14. TransformerEncoderLayer model · level 1 · ran · 81cad16cd0d5efb1
  15. _get_clones helper · level 1 · ran · our draft was wrong · 873e680b6cfcd7a5
  16. _get_activation_fn helper · level 1 · ran · our draft was wrong · d493e14665ada3f2

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2207.13820") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

3D Hand Pose Estimation3D Human Pose Estimation3D ReconstructionDecoder

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
3D Hand Pose Estimation FreiHAND FastMETRO PA-F@15mm 0.983 #16 of 33 Archive leaderboard report
3D Hand Pose Estimation FreiHAND FastMETRO PA-F@5mm 0.687 #16 of 33 Archive leaderboard report
3D Hand Pose Estimation FreiHAND FastMETRO PA-MPJPE 6.5 #16 of 33 Archive leaderboard report
3D Hand Pose Estimation FreiHAND FastMETRO PA-MPVPE 7.1 #16 of 33 Archive leaderboard report
3D Human Pose Estimation EMDB FastMETRO-L no SMPL Head Average MPJPE (mm) 108.107 #6 of 13 Archive leaderboard report
3D Human Pose Estimation EMDB FastMETRO-L no SMPL Head Average MPJPE-PA (mm) 66.794 #6 of 13 Archive leaderboard report
3D Human Pose Estimation EMDB FastMETRO-L no SMPL Head Average MVE (mm) 119.23 #6 of 13 Archive leaderboard report
3D Human Pose Estimation EMDB FastMETRO-L no SMPL Head Average MVE-PA (mm) 81.1847 #6 of 13 Archive leaderboard report
3D Human Pose Estimation EMDB FastMETRO-L no SMPL Head Jitter (10m/s^3) 185.933 #6 of 13 Archive leaderboard report
3D Human Pose Estimation EMDB FastMETRO-L Average MPJAE (deg) 25.07 #10 of 13 Archive leaderboard report
3D Human Pose Estimation EMDB FastMETRO-L Average MPJAE-PA (deg) 22.9482 #10 of 13 Archive leaderboard report
3D Human Pose Estimation EMDB FastMETRO-L Average MPJPE (mm) 115.036 #10 of 13 Archive leaderboard report
3D Human Pose Estimation EMDB FastMETRO-L Average MPJPE-PA (mm) 72.6765 #10 of 13 Archive leaderboard report
3D Human Pose Estimation EMDB FastMETRO-L Average MVE (mm) 133.566 #10 of 13 Archive leaderboard report
3D Human Pose Estimation EMDB FastMETRO-L Average MVE-PA (mm) 86.0043 #10 of 13 Archive leaderboard report
3D Human Pose Estimation EMDB FastMETRO-L Jitter (10m/s^3) 81.2959 #10 of 13 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Absolute Position EncodingsAttentionDense ConnectionsDropoutLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections