# Syn2Real: A New Benchmark forSynthetic-to-Real Visual Domain Adaptation

arXiv: https://arxiv.org/abs/1806.09755 · HTML: https://mlsanity.com/paper/1806.09755/

## Key facts

- Cited by 27 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Jun 2018.

## Built on by (most cited first, 27 of 27)

- [Predicting with Confidence on Unseen Distributions](https://mlsanity.com/paper/2107.03315/) (2021)
- [SPEED+: Next-Generation Dataset for Spacecraft Pose Estimation across Domain Gap](https://mlsanity.com/paper/2110.03101/) (2021)
- [STPLS3D: A Large-Scale Synthetic and Real Aerial Photogrammetry 3D Point Cloud Dataset](https://mlsanity.com/paper/2203.09065/) (2022)
- [RLSbench: Domain Adaptation Under Relaxed Label Shift](https://mlsanity.com/paper/2302.03020/) (2023)
- [Source-Free Progressive Graph Learning for Open-Set Domain Adaptation](https://mlsanity.com/paper/2202.06174/) (2022)
- [Towards Bridging the Space Domain Gap for Satellite Pose Estimation using Event Sensing](https://mlsanity.com/paper/2209.11945/) (2022)
- [(Almost) Provable Error Bounds Under Distribution Shift via Disagreement Discrepancy](https://mlsanity.com/paper/2306.00312/) (2023)
- [CAD-Estate: Large-scale CAD Model Annotation in RGB Videos](https://mlsanity.com/paper/2306.09011/) (2023)
- [LiDAR-CS Dataset: LiDAR Point Cloud Dataset with Cross-Sensors for 3D Object Detection](https://mlsanity.com/paper/2301.12515/) (2023)
- [Deep Transfer Learning for Intelligent Vehicle Perception: a Survey](https://mlsanity.com/paper/2306.15110/) (2023)
- [Progressively Select and Reject Pseudo-labelled Samples for Open-Set Domain Adaptation](https://mlsanity.com/paper/2110.12635/) (2021)
- [SIMSHIFT: A Benchmark for Adapting Neural Surrogates to Distribution Shifts](https://mlsanity.com/paper/2506.12007/) (2025)
- [Grounding Stylistic Domain Generalization with Quantitative Domain Shift Measures and Synthetic Scene Images](https://mlsanity.com/paper/2405.15961/) (2024)
- [Complementary Benefits of Contrastive Learning and Self-Training Under Distribution Shift](https://mlsanity.com/paper/2312.03318/) (2023)
- [Alice Benchmarks: Connecting Real World Re-Identification with the Synthetic](https://mlsanity.com/paper/2310.04416/) (2023)
- [A Survey on Deep Domain Adaptation and Tiny Object Detection Challenges, Techniques and Datasets](https://mlsanity.com/paper/2107.07927/) (2021)
- [Beyond Real Data: Synthetic Data through the Lens of Regularization](https://mlsanity.com/paper/2510.08095/) (2025)
- [ADeLA: Automatic Dense Labeling with Attention for Viewpoint Adaptation in Semantic Segmentation](https://mlsanity.com/paper/2107.14285/) (2021)
- [Adapt Anything: Tailor Any Image Classifiers across Domains And Categories Using Text-to-Image Diffusion Models](https://mlsanity.com/paper/2310.16573/) (2023)
- [ODD: Overlap-aware Estimation of Model Performance under Distribution Shift](https://mlsanity.com/paper/2506.14978/) (2025)
- [Training Kindai OCR with parallel textline images and self-attention feature distance-based loss](https://mlsanity.com/paper/2508.08537/) (2025)
- [PAS: Estimating the target accuracy before domain adaptation](https://mlsanity.com/paper/2604.09863/) (2026)
- [Asymmetric Transfer Hashing with Adaptive Bipartite Graph Learning](https://mlsanity.com/paper/2206.12592/) (2022)
- [Genetic Learning for Designing Sim-to-Real Data Augmentations](https://mlsanity.com/paper/2403.06786/) (2024)
- [Computer vision training dataset generation for robotic environments using Gaussian splatting](https://mlsanity.com/paper/2512.13411/) (2025)
- [Coarse-to-Fine Hierarchical Alignment for UAV-based Human Detection using Diffusion Models](https://mlsanity.com/paper/2512.13869/) (2025)
- [Towards Robust Cross-Dataset Object Detection Generalization under Domain Specificity](https://mlsanity.com/paper/2601.09497/) (2026)

## Abstract (arXiv)

Unsupervised transfer of object recognition models from synthetic to real data is an important problem with many potential applications. The challenge is how to "adapt" a model trained on simulated images so that it performs well on real-world data without any additional supervision. Unfortunately, current benchmarks for this problem are limited in size and task diversity. In this paper, we present a new large-scale benchmark called Syn2Real, which consists of a synthetic domain rendered from 3D object models and two real-image domains containing the same object categories. We define three related tasks on this benchmark: closed-set object classification, open-set object classification, and object detection. Our evaluation of multiple state-of-the-art methods reveals a large gap in adaptation performance between the easier closed-set classification task and the more difficult open-set and detection tasks. We conclude that developing adaptation methods that work well across all three tasks presents a significant future challenge for syn2real domain transfer.