PCGOD: Enhancing Object Detection With Synthetic Data for Scarce and Sensitive Computer Vision Tasks
Object detection models rely on large-scale, high-quality annotated datasets, which are often expensive, scarce, or restricted due to privacy concerns. Synthetic data generation has emerged as an alternative, yet existing approaches have limitations: generative models lack structured annotations and...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/11009168/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850175107612278784 |
|---|---|
| author | Walid Remmas Martin Lints Jaak Joonas Uudmae |
| author_facet | Walid Remmas Martin Lints Jaak Joonas Uudmae |
| author_sort | Walid Remmas |
| collection | DOAJ |
| description | Object detection models rely on large-scale, high-quality annotated datasets, which are often expensive, scarce, or restricted due to privacy concerns. Synthetic data generation has emerged as an alternative, yet existing approaches have limitations: generative models lack structured annotations and precise spatial control, while game-engine-based datasets suffer from inaccuracies due to 3D bounding box projections, limited scene diversity, and poor handling of articulated objects. We propose PCGOD, an Unreal Engine-based framework that combines photorealistic rendering with comprehensive domain randomization to bridge the synthetic-to-real (sim2real) domain gap. PCGOD employs a marker-based extremity projection method that places markers at key points on object geometries and projects only visible markers to create tight-fitting bounding boxes. For articulated objects, our approach dynamically tracks skeletal pose changes, ensuring annotations adapt to varied configurations. The framework addresses sim2real transfer through six-dimensional randomization: background environments, model textures and poses, landscape textures, weather conditions, camera perspectives, and procedural scene composition. Evaluations using YOLOv11 and Salience-DETR in an object detection task demonstrate that our marker-based approach achieves up to 41.61% improvement in annotation accuracy over conventional methods. Models trained with just 10% real data supplemented by our synthetic data achieve over 80% of the performance of models trained on 100% real data. Moreover, mixed datasets containing 25% synthetic and 75% real data outperform pure real-data training by up to 5.1%. These results confirm that our approach significantly enhances synthetic data utility for object detection, offering an effective solution for domains with limited training data availability. |
| format | Article |
| id | doaj-art-e2c0a2ee5ca343db8c4d5ec4e1643c14 |
| institution | OA Journals |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-e2c0a2ee5ca343db8c4d5ec4e1643c142025-08-20T02:19:31ZengIEEEIEEE Access2169-35362025-01-0113913259133310.1109/ACCESS.2025.357271911009168PCGOD: Enhancing Object Detection With Synthetic Data for Scarce and Sensitive Computer Vision TasksWalid Remmas0https://orcid.org/0000-0001-8690-0496Martin Lints1Jaak Joonas Uudmae2Research and Development Department, Defsecintel Solutions OÜ, Tallinn, EstoniaResearch and Development Department, Defsecintel Solutions OÜ, Tallinn, EstoniaResearch and Development Department, Defsecintel Solutions OÜ, Tallinn, EstoniaObject detection models rely on large-scale, high-quality annotated datasets, which are often expensive, scarce, or restricted due to privacy concerns. Synthetic data generation has emerged as an alternative, yet existing approaches have limitations: generative models lack structured annotations and precise spatial control, while game-engine-based datasets suffer from inaccuracies due to 3D bounding box projections, limited scene diversity, and poor handling of articulated objects. We propose PCGOD, an Unreal Engine-based framework that combines photorealistic rendering with comprehensive domain randomization to bridge the synthetic-to-real (sim2real) domain gap. PCGOD employs a marker-based extremity projection method that places markers at key points on object geometries and projects only visible markers to create tight-fitting bounding boxes. For articulated objects, our approach dynamically tracks skeletal pose changes, ensuring annotations adapt to varied configurations. The framework addresses sim2real transfer through six-dimensional randomization: background environments, model textures and poses, landscape textures, weather conditions, camera perspectives, and procedural scene composition. Evaluations using YOLOv11 and Salience-DETR in an object detection task demonstrate that our marker-based approach achieves up to 41.61% improvement in annotation accuracy over conventional methods. Models trained with just 10% real data supplemented by our synthetic data achieve over 80% of the performance of models trained on 100% real data. Moreover, mixed datasets containing 25% synthetic and 75% real data outperform pure real-data training by up to 5.1%. These results confirm that our approach significantly enhances synthetic data utility for object detection, offering an effective solution for domains with limited training data availability.https://ieeexplore.ieee.org/document/11009168/Synthetic datadata-augmentationobject detectionautomated labelingAI |
| spellingShingle | Walid Remmas Martin Lints Jaak Joonas Uudmae PCGOD: Enhancing Object Detection With Synthetic Data for Scarce and Sensitive Computer Vision Tasks IEEE Access Synthetic data data-augmentation object detection automated labeling AI |
| title | PCGOD: Enhancing Object Detection With Synthetic Data for Scarce and Sensitive Computer Vision Tasks |
| title_full | PCGOD: Enhancing Object Detection With Synthetic Data for Scarce and Sensitive Computer Vision Tasks |
| title_fullStr | PCGOD: Enhancing Object Detection With Synthetic Data for Scarce and Sensitive Computer Vision Tasks |
| title_full_unstemmed | PCGOD: Enhancing Object Detection With Synthetic Data for Scarce and Sensitive Computer Vision Tasks |
| title_short | PCGOD: Enhancing Object Detection With Synthetic Data for Scarce and Sensitive Computer Vision Tasks |
| title_sort | pcgod enhancing object detection with synthetic data for scarce and sensitive computer vision tasks |
| topic | Synthetic data data-augmentation object detection automated labeling AI |
| url | https://ieeexplore.ieee.org/document/11009168/ |
| work_keys_str_mv | AT walidremmas pcgodenhancingobjectdetectionwithsyntheticdataforscarceandsensitivecomputervisiontasks AT martinlints pcgodenhancingobjectdetectionwithsyntheticdataforscarceandsensitivecomputervisiontasks AT jaakjoonasuudmae pcgodenhancingobjectdetectionwithsyntheticdataforscarceandsensitivecomputervisiontasks |