# Generative Modeling for Small-Data Object Detection

arXiv: https://arxiv.org/abs/1910.07169 · HTML: https://mlsanity.com/paper/1910.07169/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Oct 2019.

## Built on by (most cited first, 20 of 20)

- [FASA: Feature Augmentation and Sampling Adaptation for Long-Tailed Instance Segmentation](https://mlsanity.com/paper/2102.12867/) (2021)
- [RenAIssance: A Survey into AI Text-to-Image Generation in the Era of Large Model](https://mlsanity.com/paper/2309.00810/) (2023)
- [A survey of synthetic data augmentation methods in computer vision](https://mlsanity.com/paper/2403.10075/) (2024)
- [Archangel: A Hybrid UAV-based Human Detection Benchmark with Position and Pose Metadata](https://mlsanity.com/paper/2209.00128/) (2022)
- [Hardness-Aware Scene Synthesis for Semi-Supervised 3D Object Detection](https://mlsanity.com/paper/2405.17422/) (2024)
- [Data-Efficient Learning via Minimizing Hyperspherical Energy](https://mlsanity.com/paper/2206.15204/) (2022)
- [Image Classification with Small Datasets: Overview and Benchmark](https://mlsanity.com/paper/2212.12478/) (2022)
- [Road images augmentation with synthetic traffic signs using neural networks](https://mlsanity.com/paper/2101.04927/) (2021)
- [An Effective Crop-Paste Pipeline for Few-shot Object Detection](https://mlsanity.com/paper/2302.14452/) (2023)
- [Box-based Refinement for Weakly Supervised and Unsupervised Localization Tasks](https://mlsanity.com/paper/2309.03874/) (2023)
- [MAC: ModAlity Calibration for Object Detection](https://mlsanity.com/paper/2310.09461/) (2023)
- [Revisiting Generative Adversarial Networks for Binary Semantic Segmentation on Imbalanced Datasets](https://mlsanity.com/paper/2402.02245/) (2024)
- [WasteGAN: Data Augmentation for Robotic Waste Sorting through Generative Adversarial Networks](https://mlsanity.com/paper/2409.16999/) (2024)
- [Generation For Adaption: A GAN-Based Approach for 3D Domain Adaption with Point Cloud Data](https://mlsanity.com/paper/2102.07373/) (2021)
- [Generative Residual Attention Network for Disease Detection](https://mlsanity.com/paper/2110.12984/) (2021)
- [On the Effectiveness of Neural Ensembles for Image Classification with Small Datasets](https://mlsanity.com/paper/2111.14493/) (2021)
- [A Survey on Cell Nuclei Instance Segmentation and Classification: Leveraging Context and Attention](https://mlsanity.com/paper/2407.18673/) (2024)
- [Accelerating Transposed Convolutions on FPGA-based Edge Devices](https://mlsanity.com/paper/2507.07683/) (2025)
- [Data Augmentation for High-Fidelity Generation of CAR-T/NK Immunological Synapse Images](https://mlsanity.com/paper/2602.00949/) (2026)
- [Beyond Pixel Similarity: Task-Aware Evaluation of GAN-Based Synthetic Sonar Data for Robotic Perception](https://mlsanity.com/paper/2609.18100/) (2026)

## Abstract (arXiv)

This paper explores object detection in the small data regime, where only a limited number of annotated bounding boxes are available due to data rarity and annotation expense. This is a common challenge today with machine learning being applied to many new tasks where obtaining training data is more challenging, e.g. in medical images with rare diseases that doctors sometimes only see once in their life-time. In this work we explore this problem from a generative modeling perspective by learning to generate new images with associated bounding boxes, and using these for training an object detector. We show that simply training previously proposed generative models does not yield satisfactory performance due to them optimizing for image realism rather than object detection accuracy. To this end we develop a new model with a novel unrolling mechanism that jointly optimizes the generative model and a detector such that the generated images improve the performance of the detector. We show this method outperforms the state of the art on two challenging datasets, disease detection and small data pedestrian detection, improving the average precision on NIH Chest X-ray by a relative 20% and localization accuracy by a relative 50%.