Papers › Learning Canonical Representations for Scene Graph to Image Generation

Learning Canonical Representations for Scene Graph to Image Generation

16 Dec 2019ECCV 2020 8arXiv:1912.07414archive 2025-07-28

Roei Herzig, Amir Bar, Huijuan Xu, Gal Chechik, Trevor Darrell, Amir Globerson

Generating realistic images of complex visual scenes becomes challenging when one wishes to control the structure of the generated images. Previous approaches showed that scenes with few entities can be controlled using scene graphs, but this approach struggles as the complexity of the graph (the number of objects and edges) increases. In this work, we show that one limitation of current methods is their inability to capture semantic equivalence in graphs. We present a novel model that addresses these issues by learning canonical graph representations from the data, resulting in improved image generation for complex visual scenes. Our model demonstrates improved empirical performance on large scene graphs, robustness to noise in the input scene graph, and generalization on semantically equivalent graphs. Finally, we show improved performance of the model on three different benchmarks: Visual Genome, COCO, and CLEVR.

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 17 code samples harvested from 1 repository linked to this paper: 10 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 2 where Syntology's instrument failed) · 7 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 17 samples from 1 repository, 10 ran (8 with no instrument failure; 2 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.

No sample is linked to this paper from 1 of its repositories that Syntology harvested for other papers:

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.

roeiherz/CanonicalSg2Im officialpytorchMITSyntology: 17 samples, 10 ran (2 instrument-failed) · 7 unverified report
roeiherz/AG2Video mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Compositional Video Synthesis with Action Graphs (arXiv:2006.15327; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 2 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method. 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

17 samples harvested: 10 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 2 where Syntology's instrument failed) · 7 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.

2ran · our draft was wrong
8ran
7unverified

Licence: 0 of the 17 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 roeiherz/CanonicalSg2Im. “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.

roeiherz/CanonicalSg2Im official repository 17 samples, 10 ran (2 instrument-failed) · 7 unverified
bce_loss in 22 papers sg2im/losses.py ran fingerprinted MIT (permissive) · ae92b5ed5c527664 · report fixture: signature input:Tensor/1/float, target:Tensor/1/float · returned Tensor · level recorded 2026-09-25
build_cnn in 7 papers sg2im/layers.py ran MIT (permissive) · d03b87199703534d · report fixture: signature arch:list/1/str, normalization:str/0/str, activation:str/0/str, padding:str/0/str, pooling:str/0/str, init:str/0/str · returned tuple · level recorded 2026-09-25
build_mlp in 74 papers sg2im/layers.py ran · our draft was wrong MIT (permissive) · a193f189a7547b8b · report fixture: signature dim_list:list/1/int, activation:str/0/str, batch_norm:str/0/str, dropout:float/0/float, final_nonlinearity:str/0/str · returned Sequential · level recorded 2026-09-22
calculate_frechet_distance in 119 papers evaluation/fid_tf.py ran MIT (permissive) · f82d43c86ff154eb · report fixture: signature mu1:ndarray/1/float, sigma1:ndarray/2/float, mu2:ndarray/1/float, sigma2:ndarray/2/float, eps:float/0/float · returned float64 · level recorded 2026-09-25
confusion_matrix in 20 papers evaluation/fid.py ran MIT (permissive) · 590d417a313b8256 · report fixture: signature num_classes:int/0/int, pred_cat:ndarray/1/int, gt_cat:ndarray/1/int · returned ndarray · level recorded 2026-09-25
gan_g_loss sg2im/losses.py ran fingerprinted MIT (permissive) · bea7eab39b0f3e75 · report fixture: signature scores_fake:Tensor/1/float · returned Tensor · level recorded 2026-09-25
get_activation in 218 papers sg2im/layers.py ran · our draft was wrong MIT (permissive) · 6adb3b06f471b25d · report fixture: signature name:str/0/str · returned ReLU · level recorded 2026-09-22
get_conv_converse sg2im/model.py ran MIT (permissive) · b117e70083ba7cf4 · report fixture: signature model:callable/0/? · returned Tensor · level recorded 2026-09-25
get_gan_losses sg2im/losses.py ran MIT (permissive) · ee94d6adfea70511 · report fixture: signature gan_type:str/0/str · returned tuple · level recorded 2026-09-25
get_predicates_weights sg2im/graph.py ran MIT (permissive) · c1b8c34d251f28dd · report fixture: signature num_preds:int/0/int, learned_init:str/0/str · returned Parameter · level recorded 2026-09-25
calculate_activation_statistics in 75 papers evaluation/fid_tf.py unverified MIT (permissive) · 6917ffcb1214d74f · report
calculate_activation_statistics in 75 papers evaluation/fid/fid_score.py unverified MIT (permissive) · 94ea5856a1451de8 · report
calculate_frechet_distance in 119 papers evaluation/fid/fid_score.py unverified MIT (permissive) · 0def50a351111624 · report
get_activations in 100 papers evaluation/fid_tf.py unverified MIT (permissive) · e69fcd6206886931 · report
get_activations in 100 papers evaluation/fid/fid_score.py unverified MIT (permissive) · 868290905012e985 · report
kl in 20 papers evaluation/fid.py unverified MIT (permissive) · df860a11dd4c724c · report
quant in 6 papers evaluation/fid.py unverified MIT (permissive) · 83ba89f8767e1931 · report

Tasks

Image GenerationLayout-to-Image GenerationScene Generation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Layout-to-Image Generation COCO-Stuff 256x256 AttSPADE FID 54.7 #5 of 5 Archive leaderboard report
Layout-to-Image Generation COCO-Stuff 256x256 AttSPADE Inception Score 15.6 #5 of 5 Archive leaderboard report
Layout-to-Image Generation COCO-Stuff 256x256 AttSPADE LPIPS 0.44 #5 of 5 Archive leaderboard report
Layout-to-Image Generation Visual Genome 256x256 AttSPADE FID 36.4 #3 of 4 Archive leaderboard report
Layout-to-Image Generation Visual Genome 256x256 AttSPADE Inception Score 11 #3 of 4 Archive leaderboard report
Layout-to-Image Generation Visual Genome 256x256 AttSPADE LPIPS 0.51 #3 of 4 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