Papers › Recovering the Unbiased Scene Graphs from the Biased Ones

Recovering the Unbiased Scene Graphs from the Biased Ones

5 Jul 2021arXiv:2107.02112archive 2025-07-28

Meng-Jiun Chiou, Henghui Ding, Hanshu Yan, Changhu Wang, Roger Zimmermann, Jiashi Feng

Given input images, scene graph generation (SGG) aims to produce comprehensive, graphical representations describing visual relationships among salient objects. Recently, more efforts have been paid to the long tail problem in SGG; however, the imbalance in the fraction of missing labels of different classes, or reporting bias, exacerbating the long tail is rarely considered and cannot be solved by the existing debiasing methods. In this paper we show that, due to the missing labels, SGG can be viewed as a "Learning from Positive and Unlabeled data" (PU learning) problem, where the reporting bias can be removed by recovering the unbiased probabilities from the biased ones by utilizing label frequencies, i.e., the per-class fraction of labeled, positive examples in all the positive examples. To obtain accurate label frequency estimates, we propose Dynamic Label Frequency Estimation (DLFE) to take advantage of training-time data augmentation and average over multiple training iterations to introduce more valid examples. Extensive experiments show that DLFE is more effective in estimating label frequencies than a naive variant of the traditional estimate, and DLFE significantly alleviates the long tail and achieves state-of-the-art debiasing performance on the VG dataset. We also show qualitatively that SGG models with DLFE produce prominently more balanced and unbiased scene graphs.

PaperPDFCode

In Syntology View this paper on Syntology, its page in Syntology's graph. That page 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

coldmanck/recovering-unbiased-scene-graphs officialmentioned in papermentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Unbiased Scene Graph Generation from Biased Training (arXiv:2002.11949; that paper's own run record: 6 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; 4 where Syntology's instrument failed) · 0 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

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Missing LabelsScene Graph ClassificationScene Graph DetectionScene Graph GenerationUnbiased Scene Graph GenerationVisual Relationship Detection

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Scene Graph Generation Visual Genome DLFE Recall@50 25.4 #10 of 19 Archive leaderboard report
Unbiased Scene Graph Generation Visual Genome DLFE (MOTIFS-ResNeXt-101-FPN backbone; PredCls mode) F@100 37.6 #3 of 31 Archive leaderboard report
Unbiased Scene Graph Generation Visual Genome DLFE (MOTIFS-ResNeXt-101-FPN backbone; PredCls mode) mR@20 22.1 #3 of 31 Archive leaderboard report
Unbiased Scene Graph Generation Visual Genome DLFE (MOTIFS-ResNeXt-101-FPN backbone; PredCls mode) ng-mR@20 30.0 #3 of 31 Archive leaderboard report
Unbiased Scene Graph Generation Visual Genome DLFE (VCTree-ResNeXt-101-FPN backbone; PredCls mode) F@100 36.0 #4 of 31 Archive leaderboard report
Unbiased Scene Graph Generation Visual Genome DLFE (VCTree-ResNeXt-101-FPN backbone; PredCls mode) mR@20 20.8 #4 of 31 Archive leaderboard report
Unbiased Scene Graph Generation Visual Genome DLFE (VCTree-ResNeXt-101-FPN backbone; PredCls mode) ng-mR@20 29.1 #4 of 31 Archive leaderboard report
Unbiased Scene Graph Generation Visual Genome DLFE (VCTree-ResNeXt-101-FPN backbone; SGCls mode) mR@20 15.8 #8 of 31 Archive leaderboard report
Unbiased Scene Graph Generation Visual Genome DLFE (VCTree-ResNeXt-101-FPN backbone; SGCls mode) ng-mR@20 21.6 #8 of 31 Archive leaderboard report
Unbiased Scene Graph Generation Visual Genome DLFE (MOTIFS-ResNeXt-101-FPN backbone; SGCls mode) F@100 21.5 #12 of 31 Archive leaderboard report
Unbiased Scene Graph Generation Visual Genome DLFE (MOTIFS-ResNeXt-101-FPN backbone; SGCls mode) mR@20 12.8 #12 of 31 Archive leaderboard report
Unbiased Scene Graph Generation Visual Genome DLFE (MOTIFS-ResNeXt-101-FPN backbone; SGCls mode) ng-mR@20 17.6 #12 of 31 Archive leaderboard report
Unbiased Scene Graph Generation Visual Genome DLFE (VCTree-ResNeXt-101-FPN backbone; SGDet mode) F@100 18.1 #17 of 31 Archive leaderboard report
Unbiased Scene Graph Generation Visual Genome DLFE (VCTree-ResNeXt-101-FPN backbone; SGDet mode) mR@20 8.6 #17 of 31 Archive leaderboard report
Unbiased Scene Graph Generation Visual Genome DLFE (VCTree-ResNeXt-101-FPN backbone; SGDet mode) ng-mR@20 11.8 #17 of 31 Archive leaderboard report
Unbiased Scene Graph Generation Visual Genome DLFE (MOTIFS-ResNeXt-101-FPN backbone; SGDet mode) F@100 18.8 #18 of 31 Archive leaderboard report
Unbiased Scene Graph Generation Visual Genome DLFE (MOTIFS-ResNeXt-101-FPN backbone; SGDet mode) mR@20 8.6 #18 of 31 Archive leaderboard report
Unbiased Scene Graph Generation Visual Genome DLFE (MOTIFS-ResNeXt-101-FPN backbone; SGDet mode) ng-mR@20 11.7 #18 of 31 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