# Improving filling level classification with adversarial training

arXiv: https://arxiv.org/abs/2102.04057 · HTML: https://mlsanity.com/paper/2102.04057/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 32 references that are in the map.
- First posted on arXiv 8 Feb 2021.

## Builds on (most cited first, 22 of 22)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Mask R-CNN](https://mlsanity.com/paper/1703.06870/) (2017)
- [Towards Deep Learning Models Resistant to Adversarial Attacks](https://mlsanity.com/paper/1706.06083/) (2017)
- [Intriguing properties of neural networks](https://mlsanity.com/paper/1312.6199/) (2013)
- [Wide Residual Networks](https://mlsanity.com/paper/1605.07146/) (2016)
- [How transferable are features in deep neural networks?](https://mlsanity.com/paper/1411.1792/) (2014)
- [Robustness May Be at Odds with Accuracy](https://mlsanity.com/paper/1805.12152/) (2018)
- [Do Better ImageNet Models Transfer Better?](https://mlsanity.com/paper/1805.08974/) (2018)
- [A Survey on Deep Transfer Learning](https://mlsanity.com/paper/1808.01974/) (2018)
- [Do Adversarially Robust ImageNet Models Transfer Better?](https://mlsanity.com/paper/2007.08489/) (2020)
- [What makes ImageNet good for transfer learning?](https://mlsanity.com/paper/1608.08614/) (2016)
- [ClearGrasp: 3D Shape Estimation of Transparent Objects for Manipulation](https://mlsanity.com/paper/1910.02550/) (2019)
- [Feature Purification: How Adversarial Training Performs Robust Deep Learning](https://mlsanity.com/paper/2005.10190/) (2020)
- [Deep Texture Manifold for Ground Terrain Recognition](https://mlsanity.com/paper/1803.10896/) (2018)
- [Optimism in the Face of Adversity: Understanding and Improving Deep Learning through Adversarial Robustness](https://mlsanity.com/paper/2010.09624/) (2020)
- [Visual Closed-Loop Control for Pouring Liquids](https://mlsanity.com/paper/1610.02610/) (2016)
- [Adversarially-Trained Deep Nets Transfer Better: Illustration on Image Classification](https://mlsanity.com/paper/2007.05869/) (2020)
- [See the Glass Half Full: Reasoning about Liquid Containers, their Volume and Content](https://mlsanity.com/paper/1701.02718/) (2017)
- [Accurate Pouring with an Autonomous Robot Using an RGB-D Camera](https://mlsanity.com/paper/1810.03303/) (2018)
- [Reasoning About Liquids via Closed-Loop Simulation](https://mlsanity.com/paper/1703.01656/) (2017)

## Built on by (most cited first, 5 of 5)

- [The CORSMAL benchmark for the prediction of the properties of containers](https://mlsanity.com/paper/2107.12719/) (2021)
- [Container Localisation and Mass Estimation with an RGB-D Camera](https://mlsanity.com/paper/2203.01207/) (2022)
- [Audio-Visual Object Classification for Human-Robot Collaboration](https://mlsanity.com/paper/2203.01977/) (2022)
- [Learning Privacy from Visual Entities](https://mlsanity.com/paper/2503.12464/) (2025)
- [Data augmentation with mixtures of max-entropy transformations for filling-level classification](https://mlsanity.com/paper/2203.04027/) (2022)

## Abstract (arXiv)

We investigate the problem of classifying - from a single image - the level of content in a cup or a drinking glass. This problem is made challenging by several ambiguities caused by transparencies, shape variations and partial occlusions, and by the availability of only small training datasets. In this paper, we tackle this problem with an appropriate strategy for transfer learning. Specifically, we use adversarial training in a generic source dataset and then refine the training with a task-specific dataset. We also discuss and experimentally evaluate several training strategies and their combination on a range of container types of the CORSMAL Containers Manipulation dataset. We show that transfer learning with adversarial training in the source domain consistently improves the classification accuracy on the test set and limits the overfitting of the classifier to specific features of the training data.