Papers › One-Stage 3D Whole-Body Mesh Recovery with Component Aware Transformer

One-Stage 3D Whole-Body Mesh Recovery with Component Aware Transformer

28 Mar 2023CVPR 2023 1arXiv:2303.16160archive 2025-07-28

Jing Lin, Ailing Zeng, Haoqian Wang, Lei Zhang, Yu Li

Whole-body mesh recovery aims to estimate the 3D human body, face, and hands parameters from a single image. It is challenging to perform this task with a single network due to resolution issues, i.e., the face and hands are usually located in extremely small regions. Existing works usually detect hands and faces, enlarge their resolution to feed in a specific network to predict the parameter, and finally fuse the results. While this copy-paste pipeline can capture the fine-grained details of the face and hands, the connections between different parts cannot be easily recovered in late fusion, leading to implausible 3D rotation and unnatural pose. In this work, we propose a one-stage pipeline for expressive whole-body mesh recovery, named OSX, without separate networks for each part. Specifically, we design a Component Aware Transformer (CAT) composed of a global body encoder and a local face/hand decoder. The encoder predicts the body parameters and provides a high-quality feature map for the decoder, which performs a feature-level upsample-crop scheme to extract high-resolution part-specific features and adopt keypoint-guided deformable attention to estimate hand and face precisely. The whole pipeline is simple yet effective without any manual post-processing and naturally avoids implausible prediction. Comprehensive experiments demonstrate the effectiveness of OSX. Lastly, we build a large-scale Upper-Body dataset (UBody) with high-quality 2D and 3D whole-body annotations. It contains persons with partially visible bodies in diverse real-life scenarios to bridge the gap between the basic task and downstream applications.

PaperPDFConference PDFCodeCode Syntology ran

In Syntology View this paper on Syntology: its repositories, every harvested function with whether it ran, its licence and the call to fetch it.

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

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2303.16160")

Code

Syntology Ran 3 of 3 code samples harvested from 1 repository linked to this paper; 0 have no recorded run. Of those that ran: 3 ran with no contract checked.

By repository: official repository: 3 samples from 1 repository, 3 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

IDEA-Research/OSX officialmentioned on GitHubpytorchMIT 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

3 samples harvested; 3 ran; 0 honoured the contract we drafted; 0 have no recorded run. Read from Syntology's graph 2026-09-25; that is when this build read the record, not when the samples ran.

3ran

Licence: 0 of the 3 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 IDEA-Research/OSX. “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. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (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 from a repository marked official are listed first. 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 four 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; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). 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.

make_conv_layers IDEA-Research/OSX/common/nets/layer.py official repository ran MIT (permissive) · 622140af456ff5d2 · report
make_deconv_layers IDEA-Research/OSX/common/nets/layer.py official repository ran MIT (permissive) · 3a26e5305cd9fa2b · report
make_linear_layers IDEA-Research/OSX/common/nets/layer.py official repository ran MIT (permissive) · 6309528f3f76d031 · report

Tasks

3D Human Pose Estimation3D Human Reconstruction3D Multi-Person Mesh RecoveryDecoder

Datasets

Introduced by this paper, per the archive.

UBody

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
3D Human Pose Estimation 3DPW OSX MPJPE 74.7 #68 of 119 Archive leaderboard report
3D Human Pose Estimation 3DPW OSX PA-MPJPE 45.1 #68 of 119 Archive leaderboard report
3D Human Pose Estimation UBody OSX PA-PVE-All 42.2 #3 of 4 Archive leaderboard report
3D Human Pose Estimation UBody OSX PA-PVE-Face 2.0 #3 of 4 Archive leaderboard report
3D Human Pose Estimation UBody OSX PA-PVE-Hands 8.6 #3 of 4 Archive leaderboard report
3D Human Pose Estimation UBody OSX PVE-All 81.9 #3 of 4 Archive leaderboard report
3D Human Pose Estimation UBody OSX PVE-Face 21.2 #3 of 4 Archive leaderboard report
3D Human Pose Estimation UBody OSX PVE-Hands 41.5 #3 of 4 Archive leaderboard report
3D Human Reconstruction EHF OSX MPVPE 70.8 #3 of 3 Archive leaderboard report
3D Human Reconstruction EHF OSX PA V2V (mm), face 6 #3 of 3 Archive leaderboard report
3D Human Reconstruction EHF OSX PA V2V (mm), whole body 48.7 #3 of 3 Archive leaderboard report
3D Multi-Person Mesh Recovery AGORA OSX B-NMVE 85.3 #4 of 7 Archive leaderboard report
3D Multi-Person Mesh Recovery AGORA OSX F-MVE 36.2 #4 of 7 Archive leaderboard report
3D Multi-Person Mesh Recovery AGORA OSX FB-MVE 122.8 #4 of 7 Archive leaderboard report
3D Multi-Person Mesh Recovery AGORA OSX FB-NMVE 130.6 #4 of 7 Archive leaderboard report
3D Multi-Person Mesh Recovery AGORA OSX LH/RH-MVE 45.7 #4 of 7 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

AWAREAbsolute Position EncodingsAdamAttentionBPECopy-PasteDense ConnectionsDropoutLabel SmoothingLayer 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