Papers › HumanBench: Towards General Human-centric Perception with Projector Assisted Pretraining

HumanBench: Towards General Human-centric Perception with Projector Assisted Pretraining

10 Mar 2023CVPR 2023 1arXiv:2303.05675archive 2025-07-28

Shixiang Tang, Cheng Chen, Qingsong Xie, Meilin Chen, Yizhou Wang, Yuanzheng Ci, Lei Bai, Feng Zhu, Haiyang Yang, Li Yi, Rui Zhao, Wanli Ouyang

Human-centric perceptions include a variety of vision tasks, which have widespread industrial applications, including surveillance, autonomous driving, and the metaverse. It is desirable to have a general pretrain model for versatile human-centric downstream tasks. This paper forges ahead along this path from the aspects of both benchmark and pretraining methods. Specifically, we propose a HumanBench based on existing datasets to comprehensively evaluate on the common ground the generalization abilities of different pretraining methods on 19 datasets from 6 diverse downstream tasks, including person ReID, pose estimation, human parsing, pedestrian attribute recognition, pedestrian detection, and crowd counting. To learn both coarse-grained and fine-grained knowledge in human bodies, we further propose a Projector AssisTed Hierarchical pretraining method (PATH) to learn diverse knowledge at different granularity levels. Comprehensive evaluations on HumanBench show that our PATH achieves new state-of-the-art results on 17 downstream datasets and on-par results on the other 2 datasets. The code will be publicly at https://github.com/OpenGVLab/HumanBench.

PaperPDFConference PDFCodeCode 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 25 code samples harvested from 1 repository linked to this paper: 15 ran (of which 1 constructed an object rather than computing a result; 8 with no instrument failure: 1 honoured, 0 violated, 7 with no contract checked; 7 where Syntology's instrument failed) · 10 unverified. Where Syntology's instrument failed: 6 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: 25 samples from 1 repository, 15 ran (8 with no instrument failure; 7 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 on 2026-09-22 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

opengvlab/humanbench officialmentioned in paperpytorchMITSyntology: 25 samples, 15 ran (7 instrument-failed) · 10 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

25 samples harvested: 15 ran (of which 1 constructed an object rather than computing a result; 8 with no instrument failure: 1 honoured, 0 violated, 7 with no contract checked; 7 where Syntology's instrument failed) · 10 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.

1ran · honoured contract
6ran · our draft was wrong
1ran; the contract check could not run
7ran
10unverified

Licence: 0 of the 25 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 opengvlab/humanbench. “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”. “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.

opengvlab/humanbench official repository 25 samples, 15 ran (7 instrument-failed) · 10 unverified
NestedTensor in 24 papers PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py ran · metamorphic tier: deterministic MIT (permissive) · 39af89396152713c · report fixture: signature device:str/0/str · returned NestedTensor · level recorded 2026-09-22
PatchEmbed in 123 papers
census: 9 implementations from 9 papers, 6 distinct outputs on one shared input
PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py ran · metamorphic tier: deterministic MIT (permissive) · 9af8e85813684a76 · report fixture: signature x:Tensor/4/float, mask:Tensor/3/float · returned tuple · level recorded 2026-09-22
autocast in 16 papers PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py ran · our draft was wrong MIT (permissive) · 69b9d45d322d1680 · report fixture: signature enabled:bool/0/bool · returned _GeneratorContextManager · level recorded 2026-09-22
enable_checkpointing PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py ran MIT (permissive) · 68e7a844f6a217c9 · report fixture: returned _GeneratorContextManager · level recorded 2026-09-22
enable_recomputing PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py ran MIT (permissive) · d32957ea95e1f955 · report fixture: returned _GeneratorContextManager · level recorded 2026-09-22
get_2d_sincos_pos_embed in 170 papers PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py ran · honoured contract fingerprinted MIT (permissive) · 0973dbf09d2066c7 · report fixture: signature embed_dim:int/0/int, grid_size:int/0/int, cls_token:bool/0/bool · returned ndarray · level recorded 2026-09-22
get_abs_pos in 15 papers PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py ran; the contract check could not run (RAISES) MIT (permissive) · 8dc7dc2bedef1191 · report fixture: signature abs_pos:Tensor/3/float, has_cls_token:bool/0/bool, hw:tuple/1/int · returned Tensor · level recorded 2026-09-22
is_autocast_enabled PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py ran MIT (permissive) · a88ee7f6d338ad8d · report fixture: returned bool · level recorded 2026-09-22
is_checkpointing PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py ran MIT (permissive) · dfb3581e62456ff2 · report fixture: returned bool · level recorded 2026-09-22
is_recomputing PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py ran MIT (permissive) · 60a3f862716c1212 · report fixture: returned bool · level recorded 2026-09-22
pack_kwargs PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py ran · our draft was wrong MIT (permissive) · 87cadb13cda1c63e · report fixture: signature args:tuple/1/int, kwargs:dict/0/? · returned tuple · level recorded 2026-09-22
patch_batchnorm PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py ran · our draft was wrong MIT (permissive) · 37dc7bc4f8eea14d · report fixture: signature module:callable/0/? · returned list · level recorded 2026-09-22
split_non_tensors PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py ran · our draft was wrong fingerprinted MIT (permissive) · c0b365b386a2ce5a · report fixture: signature mixed:Tensor/2/float · returned tuple · level recorded 2026-09-22
unpack_kwargs PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py ran · our draft was wrong fingerprinted MIT (permissive) · c577116b3b35c1c3 · report fixture: signature kwarg_keys:tuple/1/str, flat_args:tuple/1/int · returned tuple · level recorded 2026-09-22
unpack_non_tensors PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py ran · our draft was wrong MIT (permissive) · e9cfd29d0d37ed43 · report fixture: signature tensors:tuple/1/Tensor, packed_non_tensors:dict/0/? · returned tuple · level recorded 2026-09-22
CheckpointFunction in 6 papers PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py unverified MIT (permissive) · 8e99c11a4e7efdd5 · report
ViT in 10 papers PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py unverified MIT (permissive) · c8bcf97bcffa2bfd · report
WindowAttention in 50 papers PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py unverified MIT (permissive) · fc286c766c71de79 · report
_checkpointed_forward PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py unverified MIT (permissive) · 839f42723dc179f9 · report
calc_rel_pos_spatial in 5 papers PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py unverified MIT (permissive) · 37c5cac4e0bae684 · report
checkpoint_wrapper PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py unverified MIT (permissive) · 5de1a5737d6af117 · report
get_rng_state PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py unverified MIT (permissive) · f6675e1e59cce43a · report
set_rng_state PATH/core/models/backbones/vitdet_for_ladder_attention_share_pos_embed.py unverified MIT (permissive) · 9deeaa92c6fb18e5 · 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 Projector AssisTed Hierarchical pretraining method (PATH) (graph method: Projector AssisTed Hierarchical pretraining method (PATH)). 32 bricks; individually on a fixture: 21 ran (of which 3 constructed an object rather than computing a result; 11 with no instrument failure: 1 honoured, 0 violated, 10 with no contract checked; 10 where Syntology's instrument failed) · 11 unverified. The assembly has not been run end-to-end.

  1. ViT entry · level 0 · unverified · c8bcf97bcffa2bfd
  2. Block model · level 0 · unverified · 344d0eb15f2cb514
  3. NestedTensor helper · level 1 · ran · 39af89396152713c
  4. PatchEmbed model · level 1 · ran · 9af8e85813684a76
  5. QuickGELU helper · level 1 · ran · 990ae00bb84c8cce
  6. checkpoint_wrapper helper · level 0 · unverified · 5de1a5737d6af117
  7. get_2d_sincos_pos_embed helper · level 1 · ran · honoured contract · 0973dbf09d2066c7
  8. get_abs_pos model · level 1 · ran; the contract check could not run (RAISES) · 8dc7dc2bedef1191
  9. Attention model · level 0 · unverified · d2b8be1e94eca236
  10. DropPath helper · level 1 · ran · cd0e7be073ad785f
  11. Mlp helper · level 1 · ran · e71525ebf968c4c3
  12. WindowAttention model · level 0 · unverified · fc286c766c71de79
  13. _checkpointed_forward helper · level 0 · unverified · 839f42723dc179f9
  14. patch_batchnorm helper · level 1 · ran · our draft was wrong · 37dc7bc4f8eea14d
  15. get_2d_sincos_pos_embed_from_grid helper · level 0 · unverified · 2884deadbb637c3e
  16. calc_rel_pos_spatial model · level 0 · unverified · 37c5cac4e0bae684
  17. window_partition model · level 1 · ran; the contract check could not run · f9fd6241d935f07b
  18. window_reverse model · level 1 · ran · our draft was wrong · fb32094c6dbece71
  19. CheckpointFunction helper · level 0 · unverified · 8e99c11a4e7efdd5
  20. pack_kwargs helper · level 1 · ran · our draft was wrong · 87cadb13cda1c63e
  21. unpack_non_tensors helper · level 1 · ran · our draft was wrong · e9cfd29d0d37ed43
  22. get_1d_sincos_pos_embed_from_grid helper · level 1 · ran; the contract check could not run · 5c8ea92d43ead77d
  23. autocast helper · level 1 · ran · our draft was wrong · 69b9d45d322d1680
  24. enable_checkpointing helper · level 1 · ran · 68e7a844f6a217c9
  25. enable_recomputing helper · level 1 · ran · d32957ea95e1f955
  26. get_rng_state helper · level 0 · unverified · f6675e1e59cce43a
  27. is_autocast_enabled helper · level 1 · ran · a88ee7f6d338ad8d
  28. set_rng_state helper · level 0 · unverified · 9deeaa92c6fb18e5
  29. split_non_tensors helper · level 1 · ran · our draft was wrong · c0b365b386a2ce5a
  30. unpack_kwargs helper · level 1 · ran · our draft was wrong · c577116b3b35c1c3
  31. is_checkpointing helper · level 1 · ran · dfb3581e62456ff2
  32. is_recomputing helper · level 1 · ran · 60a3f862716c1212

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="2303.05675") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

AttributeAutonomous DrivingCrowd CountingHuman ParsingPedestrian Attribute RecognitionPedestrian DetectionPose Estimation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Pedestrian Attribute Recognition PA-100K PATH (Partial FT) Accuracy 90.8 #1 of 13 Archive leaderboard report
Pose Estimation COCO (Common Objects in Context) PATH (Partial FT) AP 77.1 #7 of 10 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.

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