# Realistically distributing object placements in synthetic training data improves the performance of vision-based object detection models

arXiv: https://arxiv.org/abs/2305.14621 · HTML: https://mlsanity.com/paper/2305.14621/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 23 references that are in the map.
- First posted on arXiv 24 May 2023.

## Builds on (most cited first, 16 of 16)

- [CARLA: An Open Urban Driving Simulator](https://mlsanity.com/paper/1711.03938/) (2017)
- [Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real World](https://mlsanity.com/paper/1703.06907/) (2017)
- [Training Deep Networks with Synthetic Data: Bridging the Reality Gap by Domain Randomization](https://mlsanity.com/paper/1804.06516/) (2018)
- [CAD2RL: Real Single-Image Flight without a Single Real Image](https://mlsanity.com/paper/1611.04201/) (2016)
- [Probabilistic and Geometric Depth: Detecting Objects in Perspective](https://mlsanity.com/paper/2107.14160/) (2021)
- [Meta-Sim: Learning to Generate Synthetic Datasets](https://mlsanity.com/paper/1904.11621/) (2019)
- [LayoutVAE: Stochastic Scene Layout Generation From a Label Set](https://mlsanity.com/paper/1907.10719/) (2019)
- [Learning to Track with Object Permanence](https://mlsanity.com/paper/2103.14258/) (2021)
- [Structured Domain Randomization: Bridging the Reality Gap by Context-Aware Synthetic Data](https://mlsanity.com/paper/1810.10093/) (2018)
- [SceneGen: Learning to Generate Realistic Traffic Scenes](https://mlsanity.com/paper/2101.06541/) (2021)
- [Meta-Sim2: Unsupervised Learning of Scene Structure for Synthetic Data Generation](https://mlsanity.com/paper/2008.09092/) (2020)
- [Synthesizing Training Data for Object Detection in Indoor Scenes](https://mlsanity.com/paper/1702.07836/) (2017)
- [How much real data do we actually need: Analyzing object detection performance using synthetic and real data](https://mlsanity.com/paper/1907.07061/) (2019)
- [Domain Randomization for Scene-Specific Car Detection and Pose Estimation](https://mlsanity.com/paper/1811.05939/) (2018)
- [Self-Supervised Real-to-Sim Scene Generation](https://mlsanity.com/paper/2011.14488/) (2020)
- [Conditional Permutation Invariant Flows](https://mlsanity.com/paper/2206.09021/) (2022)

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

- [Prior Distribution and Model Confidence](https://mlsanity.com/paper/2509.05485/) (2025)

## Abstract (arXiv)

When training object detection models on synthetic data, it is important to make the distribution of synthetic data as close as possible to the distribution of real data. We investigate specifically the impact of object placement distribution, keeping all other aspects of synthetic data fixed. Our experiment, training a 3D vehicle detection model in CARLA and testing on KITTI, demonstrates a substantial improvement resulting from improving the object placement distribution.