# Joint Training of Image Generator and Detector for Road Defect Detection

arXiv: https://arxiv.org/abs/2509.03465 · HTML: https://mlsanity.com/paper/2509.03465/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 24 references that are in the map.
- First posted on arXiv 3 Sep 2025.

## Builds on (most cited first, 10 of 10)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [InternImage: Exploring Large-Scale Vision Foundation Models with Deformable Convolutions](https://mlsanity.com/paper/2211.05778/) (2022)
- [The Role of ImageNet Classes in Fr'echet Inception Distance](https://mlsanity.com/paper/2203.06026/) (2022)
- [YOLOP: You Only Look Once for Panoptic Driving Perception](https://mlsanity.com/paper/2108.11250/) (2021)
- [RDD2022: A multi-national image dataset for automatic Road Damage Detection](https://mlsanity.com/paper/2209.08538/) (2022)
- [Ten Years of Generative Adversarial Nets (GANs): A survey of the state-of-the-art](https://mlsanity.com/paper/2308.16316/) (2023)
- [Crowdsensing-based Road Damage Detection Challenge (CRDDC-2022)](https://mlsanity.com/paper/2211.11362/) (2022)
- [Iterative Self Knowledge Distillation -- From Pothole Classification to Fine-Grained and COVID Recognition](https://mlsanity.com/paper/2202.02265/) (2022)

## Abstract (arXiv)

Road defect detection is important for road authorities to reduce the vehicle damage caused by road defects. Considering the practical scenarios where the defect detectors are typically deployed on edge devices with limited memory and computational resource, we aim at performing road defect detection without using ensemble-based methods or test-time augmentation (TTA). To this end, we propose to Jointly Train the image Generator and Detector for road defect detection (dubbed as JTGD). We design the dual discriminators for the generative model to enforce both the synthesized defect patches and overall images to look plausible. The synthesized image quality is improved by our proposed CLIP-based Fr'echet Inception Distance loss. The generative model in JTGD is trained jointly with the detector to encourage the generative model to synthesize harder examples for the detector. Since harder synthesized images of better quality caused by the aforesaid design are used in the data augmentation, JTGD outperforms the state-of-the-art method in the RDD2022 road defect detection benchmark across various countries under the condition of no ensemble and TTA. JTGD only uses less than 20% of the number of parameters compared with the competing baseline, which makes it more suitable for deployment on edge devices in practice.