Papers › 3D-Aware Visual Question Answering about Parts, Poses and Occlusions

3D-Aware Visual Question Answering about Parts, Poses and Occlusions

27 Oct 2023NeurIPS 2023 11arXiv:2310.17914archive 2025-07-28

Xingrui Wang, Wufei Ma, Zhuowan Li, Adam Kortylewski, Alan Yuille

Despite rapid progress in Visual question answering (VQA), existing datasets and models mainly focus on testing reasoning in 2D. However, it is important that VQA models also understand the 3D structure of visual scenes, for example to support tasks like navigation or manipulation. This includes an understanding of the 3D object pose, their parts and occlusions. In this work, we introduce the task of 3D-aware VQA, which focuses on challenging questions that require a compositional reasoning over the 3D structure of visual scenes. We address 3D-aware VQA from both the dataset and the model perspective. First, we introduce Super-CLEVR-3D, a compositional reasoning dataset that contains questions about object parts, their 3D poses, and occlusions. Second, we propose PO3D-VQA, a 3D-aware VQA model that marries two powerful ideas: probabilistic neural symbolic program execution for reasoning and deep neural networks with 3D generative representations of objects for robust visual recognition. Our experimental results show our model PO3D-VQA outperforms existing methods significantly, but we still observe a significant performance gap compared to 2D VQA benchmarks, indicating that 3D-aware VQA remains an important open research area.

PaperPDFConference PDFCodeCode Syntology ran

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 28 code samples harvested from 2 repositories linked to this paper: 22 ran (of which 0 constructed an object rather than computing a result; 21 with no instrument failure: 0 honoured, 0 violated, 21 with no contract checked; 1 where Syntology's instrument failed) · 6 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 14 samples from 1 repository, 11 ran (10 with no instrument failure; 1 where Syntology's instrument failed); community (archive-listed): 14 samples from 1 repository, 11 ran (11 with no instrument failure; 0 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 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

xingruiwang/3d-aware-vqa officialmentioned in papermentioned on GitHubpytorchSyntology: 14 samples, 11 ran (1 instrument-failed) · 3 unverified report
XingruiWang/superclevr-3D-question mentioned on GitHubMITSyntology: 14 samples, 11 ran (0 instrument-failed) · 3 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

28 samples harvested: 22 ran (of which 0 constructed an object rather than computing a result; 21 with no instrument failure: 0 honoured, 0 violated, 21 with no contract checked; 1 where Syntology's instrument failed) · 6 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 · our draft was wrong
21ran
6unverified

Licence: 14 of the 28 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 2 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “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.

xingruiwang/3d-aware-vqa official repository 14 samples, 11 ran (1 instrument-failed) · 3 unverified
convert_bbox PO3D-VQA/6D_pose_estimator/align_objects.py ran no licence file found · pointer only · b3c724d540d72cce · report fixture: signature bbox:list/1/int, format:str/0/str · returned dict · level recorded 2026-09-24
direction_probabilities PO3D-VQA/6D_pose_estimator/align_objects_parts_prob.py ran fingerprinted no licence file found · pointer only · 84fe49b6fbbf5f07 · report fixture: signature degree:int/0/int · returned list · level recorded 2026-09-24
extract_bboxes in 14 papers PO3D-VQA/6D_pose_estimator/post_process.py ran no licence file found · pointer only · 46aeea7280866e13 · report fixture: signature mask:ndarray/2/int · returned list · level recorded 2026-09-24
filter_object PO3D-VQA/6D_pose_estimator/post_process.py ran no licence file found · pointer only · 2a2da93e76d3b6b1 · report fixture: signature class_objects:dict/0/?, img:str/0/str, vis:bool/0/bool · returned list · level recorded 2026-09-24
get_attrs_clevr PO3D-VQA/attr_net/utils.py ran fingerprinted no licence file found · pointer only · b24784b77cae1db3 · report fixture: signature feat_vec:ndarray/1/float · returned dict · level recorded 2026-09-24
get_iou in 34 papers PO3D-VQA/6D_pose_estimator/align_objects.py ran no licence file found · pointer only · a080ed5b2ef7aedb · report fixture: signature bb1:list/1/int, bb2:list/1/int · returned float · level recorded 2026-09-24
get_overlap in 8 papers PO3D-VQA/6D_pose_estimator/post_process.py ran fingerprinted no licence file found · pointer only · bc0a3b6b3b7d1cde · report fixture: signature seg1:ndarray/2/int, seg2:ndarray/2/int · returned float64 · level recorded 2026-09-24
get_trainer in 5 papers PO3D-VQA/attr_net/trainer.py ran no licence file found · pointer only · 712891b9264da608 · report fixture: signature opt:DummyOpt/0/?, model:callable/0/?, train_loader:list/1/tuple, val_loader:list/1/tuple · returned Trainer · level recorded 2026-09-24
invert_dict in 13 papers PO3D-VQA/attr_net/utils.py ran · our draft was wrong no licence file found · pointer only · e6dbe2d3b6f0cc87 · report fixture: signature d:dict/0/? · returned dict · level recorded 2026-09-22
one_hot in 150 papers
census: 18 implementations from 25 papers, 10 distinct outputs across 6 buckets, one shared input per bucket
PO3D-VQA/6D_pose_estimator/align_objects_parts_prob.py ran no licence file found · pointer only · 57f66e40396ced61 · report fixture: signature value:str/0/str, categories:list/1/str · returned list · level recorded 2026-09-24
process_pose PO3D-VQA/6D_pose_estimator/align_objects.py ran fingerprinted no licence file found · pointer only · 1778773a6330dd75 · report fixture: signature rotation:int/0/int · returned str · level recorded 2026-09-24
get_feat_vec_clevr PO3D-VQA/attr_net/utils.py unverified no licence file found · pointer only · 80e331ff8945a5c9 · report
get_model in 368 papers PO3D-VQA/attr_net/model-2.py unverified no licence file found · pointer only · 5a49cf40fd3eef79 · report
get_model in 368 papers PO3D-VQA/attr_net/model.py unverified no licence file found · pointer only · c23434d78dbe6b73 · report
XingruiWang/superclevr-3D-question community (archive-listed) 14 samples, 11 ran (0 instrument-failed) · 3 unverified
add_pose image_generation/add_occlusion_mask/add_occlusion.py ran MIT (permissive) · fee3bf99805ee6cb · report fixture: signature scene:dict/0/? · returned dict · level recorded 2026-09-25
align in 23 papers question_generation/align_removed.py ran MIT (permissive) · 24cf398747aafdc0 · report fixture: signature old_data:list/1/dict, new_data:list/1/dict · returned list · level recorded 2026-09-25
compute_all_relationships image_generation/super_restore_render_images.py ran MIT (permissive) · 6814863b0a1e72a9 · report fixture: signature scene_struct:dict/0/?, eps:float/0/float · returned dict · level recorded 2026-09-24
extract_args image_generation/utils.py ran fingerprinted MIT (permissive) · a3dbae9548681fea · report fixture: signature input_argv:list/1/str · returned list · level recorded 2026-09-24
get_candidates in 12 papers question_generation/align_removed.py ran MIT (permissive) · c576aa00026275ff · report fixture: signature lookup:dict/0/?, key:str/0/str, answer:str/0/str, used:set/0/? · returned list · level recorded 2026-09-24
get_mask in 119 papers
census: 9 implementations from 10 papers, 9 distinct outputs across 7 buckets, one shared input per bucket
image_generation/restore_img2json.py ran MIT (permissive) · 05df8fe432cb08a8 · report fixture: signature path:str/0/str, mat_indices:dict/0/? · returned tuple · level recorded 2026-09-24
parse_args in 355 papers image_generation/utils.py ran MIT (permissive) · 5487421f055b3459 · report fixture: signature parser:ArgumentParser/0/?, argv:list/1/str · returned Namespace · level recorded 2026-09-24
scene_handler question_generation/question_engine.py ran MIT (permissive) · 2b0321ad9cda9dcb · report fixture: signature scene_struct:dict/0/?, inputs:Tensor/2/float, side_inputs:dict/0/? · returned list · level recorded 2026-09-24
srgb_to_linear in 10 papers image_generation/restore_img2json.py ran fingerprinted MIT (permissive) · 7ad3633d3e96f006 · report fixture: signature x:float/0/float, mod:float/0/float · returned float · level recorded 2026-09-24
to_str in 10 papers image_generation/restore_img2json.py ran fingerprinted MIT (permissive) · a2ed7edd94a0f20c · report fixture: signature img:list/1/int · returned str · level recorded 2026-09-25
unique_handler question_generation/question_engine.py ran MIT (permissive) · 515aaf0b48a01dcd · report fixture: signature scene_struct:dict/0/?, inputs:list/1/list, side_inputs:dict/0/? · returned str · level recorded 2026-09-24
check_visibility image_generation/super_restore_render_images_split.py unverified MIT (permissive) · f9b74271607a3291 · report
load_properties_json image_generation/preprocess_cgpart.py unverified MIT (permissive) · a37183fd64f6eb8c · report
render_shadeless image_generation/super_restore_render_images_split.py unverified MIT (permissive) · 967a2489572384a0 · report

Tasks

Question AnsweringVisual Question AnsweringVisual Question Answering (VQA)

Datasets

Introduced by this paper, per the archive.

Super-CLEVR-3D

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Introduced by this paper: PO3D-VQA

MDETRPO3D-VQA

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