# Unbiased Scene Graph Generation from Biased Training

arXiv: https://arxiv.org/abs/2002.11949 · HTML: https://mlsanity.com/paper/2002.11949/

## Key facts

- Cited by 355 papers in the map (data through 7 Oct 2026).
- First posted on arXiv 27 Feb 2020.

## Built on by (most cited first, 30 of 355)

- [STAR: A Benchmark for Situated Reasoning in Real-World Videos](https://mlsanity.com/paper/2405.09711/) (2024)
- [Bipartite Graph Network with Adaptive Message Passing for Unbiased Scene Graph Generation](https://mlsanity.com/paper/2104.00308/) (2021)
- [A Comprehensive Survey of Scene Graphs: Generation and Application](https://mlsanity.com/paper/2104.01111/) (2021)
- [Distilling Causal Effect of Data in Class-Incremental Learning](https://mlsanity.com/paper/2103.01737/) (2021)
- [Counterfactual Attention Learning for Fine-Grained Visual Categorization and Re-identification](https://mlsanity.com/paper/2108.08728/) (2021)
- [Panoptic Scene Graph Generation](https://mlsanity.com/paper/2207.11247/) (2022)
- [Causal Attention for Vision-Language Tasks](https://mlsanity.com/paper/2103.03493/) (2021)
- [Energy-Based Learning for Scene Graph Generation](https://mlsanity.com/paper/2103.02221/) (2021)
- [SGTR: End-to-end Scene Graph Generation with Transformer](https://mlsanity.com/paper/2112.12970/) (2021)
- [SkySenseGPT: A Fine-Grained Instruction Tuning Dataset and Model for Remote Sensing Vision-Language Understanding](https://mlsanity.com/paper/2406.10100/) (2024)
- [RelTR: Relation Transformer for Scene Graph Generation](https://mlsanity.com/paper/2201.11460/) (2022)
- [Interventional Video Grounding with Dual Contrastive Learning](https://mlsanity.com/paper/2106.11013/) (2021)
- [Semantic Communications: Principles and Challenges](https://mlsanity.com/paper/2201.01389/) (2021)
- [Causal Attention for Unbiased Visual Recognition](https://mlsanity.com/paper/2108.08782/) (2021)
- [Learning and Evaluating Graph Neural Network Explanations based on Counterfactual and Factual Reasoning](https://mlsanity.com/paper/2202.08816/) (2022)
- [Multi-Modal Knowledge Graph Construction and Application: A Survey](https://mlsanity.com/paper/2202.05786/) (2022)
- [Scene Graph Generation: A Comprehensive Survey](https://mlsanity.com/paper/2201.00443/) (2022)
- [Recovering the Unbiased Scene Graphs from the Biased Ones](https://mlsanity.com/paper/2107.02112/) (2021)
- [Stacked Hybrid-Attention and Group Collaborative Learning for Unbiased Scene Graph Generation](https://mlsanity.com/paper/2203.09811/) (2022)
- [Learning of Visual Relations: The Devil is in the Tails](https://mlsanity.com/paper/2108.09668/) (2021)
- [Learning to Generate Scene Graph from Natural Language Supervision](https://mlsanity.com/paper/2109.02227/) (2021)
- [The Devil is in the Labels: Noisy Label Correction for Robust Scene Graph Generation](https://mlsanity.com/paper/2206.03014/) (2022)
- [Fully Convolutional Scene Graph Generation](https://mlsanity.com/paper/2103.16083/) (2021)
- [Fine-Grained Scene Graph Generation with Data Transfer](https://mlsanity.com/paper/2203.11654/) (2022)
- [Cross-Modal Causal Relational Reasoning for Event-Level Visual Question Answering](https://mlsanity.com/paper/2207.12647/) (2022)
- [Towards Open-vocabulary Scene Graph Generation with Prompt-based Finetuning](https://mlsanity.com/paper/2208.08165/) (2022)
- [From General to Specific: Informative Scene Graph Generation via Balance Adjustment](https://mlsanity.com/paper/2108.13129/) (2021)
- [Equivariant Similarity for Vision-Language Foundation Models](https://mlsanity.com/paper/2303.14465/) (2023)
- [FACTUAL: A Benchmark for Faithful and Consistent Textual Scene Graph Parsing](https://mlsanity.com/paper/2305.17497/) (2023)
- [EGTR: Extracting Graph from Transformer for Scene Graph Generation](https://mlsanity.com/paper/2404.02072/) (2024)

## Abstract (arXiv)

Today's scene graph generation (SGG) task is still far from practical, mainly due to the severe training bias, e.g., collapsing diverse "human walk on / sit on / lay on beach" into "human on beach". Given such SGG, the down-stream tasks such as VQA can hardly infer better scene structures than merely a bag of objects. However, debiasing in SGG is not trivial because traditional debiasing methods cannot distinguish between the good and bad bias, e.g., good context prior (e.g., "person read book" rather than "eat") and bad long-tailed bias (e.g., "near" dominating "behind / in front of"). In this paper, we present a novel SGG framework based on causal inference but not the conventional likelihood. We first build a causal graph for SGG, and perform traditional biased training with the graph. Then, we propose to draw the counterfactual causality from the trained graph to infer the effect from the bad bias, which should be removed. In particular, we use Total Direct Effect (TDE) as the proposed final predicate score for unbiased SGG. Note that our framework is agnostic to any SGG model and thus can be widely applied in the community who seeks unbiased predictions. By using the proposed Scene Graph Diagnosis toolkit on the SGG benchmark Visual Genome and several prevailing models, we observed significant improvements over the previous state-of-the-art methods.