# Curating Synthetic Data for Task-Specific Visual Perception

arXiv: https://arxiv.org/abs/2609.38476 · HTML: https://mlsanity.com/paper/2609.38476/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 15 of its 24 references that are in the map.
- First posted on arXiv 29 Sep 2026.

## Builds on (most cited first, 15 of 15)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimation](https://mlsanity.com/paper/1512.02134/) (2015)
- [Playing for Data: Ground Truth from Computer Games](https://mlsanity.com/paper/1608.02192/) (2016)
- [PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenes](https://mlsanity.com/paper/1711.00199/) (2017)
- [Inverse Rendering for Complex Indoor Scenes: Shape, Spatially-Varying Lighting and SVBRDF from a Single Image](https://mlsanity.com/paper/1905.02722/) (2019)
- [Single-Image SVBRDF Capture with a Rendering-Aware Deep Network](https://mlsanity.com/paper/1810.09718/) (2018)
- [MOTSynth: How Can Synthetic Data Help Pedestrian Detection and Tracking?](https://mlsanity.com/paper/2108.09518/) (2021)
- [3D Face Reconstruction by Learning from Synthetic Data](https://mlsanity.com/paper/1609.04387/) (2016)
- [Synthetic Datasets for Autonomous Driving: A Survey](https://mlsanity.com/paper/2304.12205/) (2023)
- [LW-DETR: A Transformer Replacement to YOLO for Real-Time Detection](https://mlsanity.com/paper/2406.03459/) (2024)
- [FRED: The Florence RGB-Event Drone Dataset](https://mlsanity.com/paper/2506.05163/) (2025)
- [6D Strawberry Pose Estimation: Real-time and Edge AI Solutions Using Purely Synthetic Training Data](https://mlsanity.com/paper/2511.11307/) (2025)
- [SPEAR: A Simulator for Photorealistic Embodied AI Research](https://mlsanity.com/paper/2607.06701/) (2026)
- [Neural Restoration of Greening Defects in Historical Autochrome Photographs Based on Purely Synthetic Data](https://mlsanity.com/paper/2505.22291/) (2025)
- [YCB-Ev SD: Synthetic event-vision dataset for 6DoF object pose estimation](https://mlsanity.com/paper/2511.11344/) (2025)

## Abstract (arXiv)

Synthetic data are most valuable where general-purpose datasets cannot provide the domain-specific priors a task requires, and where manual annotation is expensive, imprecise, or infeasible. In this article we argue that the central question for specialized vision systems is not how to generate more data, but which data to generate. We therefore discuss curated synthetic data, whose scene content, appearance variations, sensing characteristics, and annotations are deliberately designed around a given task. We examine three complementary curation paradigms. Procedural rendering offers explicit control over scene parameters and the annotations follow by construction. Physically-based simulation encodes the mechanism behind an observed effect and yields exactly aligned supervision pairs. Generative AI learns sensor-specific appearance from small real seed sets and attains plausible realism, though it remains prone to hallucination and to inaccurate annotation. These paradigms are illustrated with examples from industrial surface defect detection, restoration of degraded digitized autochrome plates, and 6DoF pose estimation from RGB and event data. Using these examples, we analyze different data regimes and training strategies that combine synthetic and real data across these paradigms. We conclude that curated synthetic data are best understood as a complement to real observations, and that hybrid pipelines combining controllable supervision with learned appearance are the most promising direction for reliable sim-to-real transfer.