# Modeling Camera Effects to Improve Visual Learning from Synthetic Data

arXiv: https://arxiv.org/abs/1803.07721 · HTML: https://mlsanity.com/paper/1803.07721/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Mar 2018.

## Built on by (most cited first, 11 of 11)

- [Automatically Prepare Training Data for YOLO Using Robotic In-Hand Observation and Synthesis](https://mlsanity.com/paper/2301.01441/) (2023)
- [VR Facial Animation for Immersive Telepresence Avatars](https://mlsanity.com/paper/2304.12051/) (2023)
- [Surround-View Fisheye Optics in Computer Vision and Simulation: Survey and Challenges](https://mlsanity.com/paper/2402.12041/) (2024)
- [Active shooter detection and robust tracking utilizing supplemental synthetic data](https://mlsanity.com/paper/2309.03381/) (2023)
- [Attention-Based VR Facial Animation with Visual Mouth Camera Guidance for Immersive Telepresence Avatars](https://mlsanity.com/paper/2312.09750/) (2023)
- [Ontology-Guided Diffusion for Zero-Shot Visual Sim2Real Transfer](https://mlsanity.com/paper/2603.18719/) (2026)
- [Comprehensive Analysis of the Object Detection Pipeline on UAVs](https://mlsanity.com/paper/2203.00306/) (2022)
- [VALERIE22 -- A photorealistic, richly metadata annotated dataset of urban environments](https://mlsanity.com/paper/2308.09632/) (2023)
- [SS-SFR: Synthetic Scenes Spatial Frequency Response on Virtual KITTI and Degraded Automotive Simulations for Object Detection](https://mlsanity.com/paper/2407.15646/) (2024)
- [Robust soybean seed yield estimation using high-throughput ground robot videos](https://mlsanity.com/paper/2412.02642/) (2024)
- [DiffPhysCam: Differentiable Physics-Based Camera Simulation for Inverse Rendering and Embodied AI](https://mlsanity.com/paper/2508.08831/) (2025)

## Abstract (arXiv)

Recent work has focused on generating synthetic imagery to increase the size and variability of training data for learning visual tasks in urban scenes. This includes increasing the occurrence of occlusions or varying environmental and weather effects. However, few have addressed modeling variation in the sensor domain. Sensor effects can degrade real images, limiting generalizability of network performance on visual tasks trained on synthetic data and tested in real environments. This paper proposes an efficient, automatic, physically-based augmentation pipeline to vary sensor effects --chromatic aberration, blur, exposure, noise, and color cast-- for synthetic imagery. In particular, this paper illustrates that augmenting synthetic training datasets with the proposed pipeline reduces the domain gap between synthetic and real domains for the task of object detection in urban driving scenes.