# Measuring the Effect of Background on Classification and Feature Importance in Deep Learning for AV Perception

arXiv: https://arxiv.org/abs/2512.05937 · HTML: https://mlsanity.com/paper/2512.05937/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 41 references that are in the map.
- First posted on arXiv 5 Dec 2025.

## Builds on (most cited first, 19 of 19)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [Playing for Data: Ground Truth from Computer Games](https://mlsanity.com/paper/1608.02192/) (2016)
- [Virtual KITTI 2](https://mlsanity.com/paper/2001.10773/) (2020)
- [Virtual Worlds as Proxy for Multi-Object Tracking Analysis](https://mlsanity.com/paper/1605.06457/) (2016)
- [From Anecdotal Evidence to Quantitative Evaluation Methods: A Systematic Review on Evaluating Explainable AI](https://mlsanity.com/paper/2201.08164/) (2022)
- [Right for the Right Reasons: Training Differentiable Models by Constraining their Explanations](https://mlsanity.com/paper/1703.03717/) (2017)
- [Driving in the Matrix: Can Virtual Worlds Replace Human-Generated Annotations for Real World Tasks?](https://mlsanity.com/paper/1610.01983/) (2016)
- [Synthetic Data for Deep Learning](https://mlsanity.com/paper/1909.11512/) (2019)
- [FunnyBirds: A Synthetic Vision Dataset for a Part-Based Analysis of Explainable AI Methods](https://mlsanity.com/paper/2308.06248/) (2023)
- [Simulating Content Consistent Vehicle Datasets with Attribute Descent](https://mlsanity.com/paper/1912.08855/) (2019)
- [CURE-TSR: Challenging Unreal and Real Environments for Traffic Sign Recognition](https://mlsanity.com/paper/1712.02463/) (2017)
- [Visual Explanation by Interpretation: Improving Visual Feedback Capabilities of Deep Neural Networks](https://mlsanity.com/paper/1712.06302/) (2017)
- [Shortcomings of Top-Down Randomization-Based Sanity Checks for Evaluations of Deep Neural Network Explanations](https://mlsanity.com/paper/2211.12486/) (2022)
- [Focus! Rating XAI Methods and Finding Biases](https://mlsanity.com/paper/2109.15035/) (2021)
- [Classification Metrics for Image Explanations: Towards Building Reliable XAI-Evaluations](https://mlsanity.com/paper/2406.05068/) (2024)
- [Deep Traffic Sign Detection and Recognition Without Target Domain Real Images](https://mlsanity.com/paper/2008.00962/) (2020)

## Built on by (most cited first, 1 of 1)

- [The Garbage Dataset (GD): A Multi-Class Image Benchmark for Automated Waste Segregation](https://mlsanity.com/paper/2602.10500/) (2026)

## Abstract (arXiv)

Common approaches to explainable AI (XAI) for deep learning focus on analyzing the importance of input features on the classification task in a given model: saliency methods like SHAP and GradCAM are used to measure the impact of spatial regions of the input image on the classification result. Combined with ground truth information about the location of the object in the input image (e.g., a binary mask), it is determined whether object pixels had a high impact on the classification result, or whether the classification focused on background pixels. The former is considered to be a sign of a healthy classifier, whereas the latter is assumed to suggest overfitting on spurious correlations. A major challenge, however, is that these intuitive interpretations are difficult to test quantitatively, and hence the output of such explanations lacks an explanation itself. One particular reason is that correlations in real-world data are difficult to avoid, and whether they are spurious or legitimate is debatable. Synthetic data in turn can facilitate to actively enable or disable correlations where desired but often lack a sufficient quantification of realism and stochastic properties. [...] Therefore, we systematically generate six synthetic datasets for the task of traffic sign recognition, which differ only in their degree of camera variation and background correlation [...] to quantify the isolated influence of background correlation, different levels of camera variation, and considered traffic sign shapes on the classification performance, as well as background feature importance. [...] Results include a quantification of when and how much background features gain importance to support the classification task based on changes in the training domain [...]. Download: synset.de/datasets/synset-signset-ger/background-effect