Synthetic image data generation via rendering techniques for training AI-based instance segmentation
Synthetic image data generation has gained popularity in computer vision and machine learning in recent years. The work introduces a technique for creating artificial image data by utilizing 3D files and rendering methods in Python and Blender. The technique employs BlenderProc, a rendering tool for...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Semarak Ilmu Publishing
2026
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/42815/ http://umpir.ump.edu.my/id/eprint/42815/1/Synthetic%20image%20data%20generation%20via%20rendering%20techniques%20for%20training%20AI-based.pdf |
| _version_ | 1848826708065517568 |
|---|---|
| author | Kho, Dickson Yik Cheng Norazlianie, Sazali Ismayuzri, Ishak Saiful Anwar, Che Ghani |
| author_facet | Kho, Dickson Yik Cheng Norazlianie, Sazali Ismayuzri, Ishak Saiful Anwar, Che Ghani |
| author_sort | Kho, Dickson Yik Cheng |
| building | UMP Institutional Repository |
| collection | Online Access |
| description | Synthetic image data generation has gained popularity in computer vision and machine learning in recent years. The work introduces a technique for creating artificial image data by utilizing 3D files and rendering methods in Python and Blender. The technique employs BlenderProc, a rendering tool for generating artificial images, to efficiently create a substantial amount of data. The output of the method is saved in JSON format, containing COCO annotations of objects in the images, facilitating seamless integration with current machine-learning pipelines. The paper shows that the created synthetic data can be used to enhance object data during simulation. The method can enhance the accuracy and robustness of machine-learning models by modifying simulation parameters like lighting, camera position, and object orientation to create a variety of images. This is especially beneficial for applications that require significant amounts of labelled real-world data, which can be time-consuming and labour-intensive to obtain. The study addresses the constraints and potential prejudices of creating synthetic data, emphasizing the significance of verifying and assessing the generated data prior to its utilization in machine learning models. Synthetic data generation can be a valuable tool for improving the efficiency and effectiveness of machine learning and computer vision applications. However, it is crucial to thoroughly assess the potential limitations and biases of the generated data. This paper emphasizes the potential of synthetic data generation to enhance the accuracy and resilience of machine learning models, especially in scenarios with limited access to labelled real-world data. This paper introduces a method that efficiently produces substantial amounts of synthetic image data with COCO annotations, serving as a valuable resource for professionals in computer vision and machine learning. |
| first_indexed | 2025-11-15T03:49:06Z |
| format | Article |
| id | ump-42815 |
| institution | Universiti Malaysia Pahang |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-15T03:49:06Z |
| publishDate | 2026 |
| publisher | Semarak Ilmu Publishing |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | ump-428152024-12-31T00:14:34Z http://umpir.ump.edu.my/id/eprint/42815/ Synthetic image data generation via rendering techniques for training AI-based instance segmentation Kho, Dickson Yik Cheng Norazlianie, Sazali Ismayuzri, Ishak Saiful Anwar, Che Ghani TJ Mechanical engineering and machinery TS Manufactures Synthetic image data generation has gained popularity in computer vision and machine learning in recent years. The work introduces a technique for creating artificial image data by utilizing 3D files and rendering methods in Python and Blender. The technique employs BlenderProc, a rendering tool for generating artificial images, to efficiently create a substantial amount of data. The output of the method is saved in JSON format, containing COCO annotations of objects in the images, facilitating seamless integration with current machine-learning pipelines. The paper shows that the created synthetic data can be used to enhance object data during simulation. The method can enhance the accuracy and robustness of machine-learning models by modifying simulation parameters like lighting, camera position, and object orientation to create a variety of images. This is especially beneficial for applications that require significant amounts of labelled real-world data, which can be time-consuming and labour-intensive to obtain. The study addresses the constraints and potential prejudices of creating synthetic data, emphasizing the significance of verifying and assessing the generated data prior to its utilization in machine learning models. Synthetic data generation can be a valuable tool for improving the efficiency and effectiveness of machine learning and computer vision applications. However, it is crucial to thoroughly assess the potential limitations and biases of the generated data. This paper emphasizes the potential of synthetic data generation to enhance the accuracy and resilience of machine learning models, especially in scenarios with limited access to labelled real-world data. This paper introduces a method that efficiently produces substantial amounts of synthetic image data with COCO annotations, serving as a valuable resource for professionals in computer vision and machine learning. Semarak Ilmu Publishing 2026 Article PeerReviewed pdf en cc_by_nc_4 http://umpir.ump.edu.my/id/eprint/42815/1/Synthetic%20image%20data%20generation%20via%20rendering%20techniques%20for%20training%20AI-based.pdf Kho, Dickson Yik Cheng and Norazlianie, Sazali and Ismayuzri, Ishak and Saiful Anwar, Che Ghani (2026) Synthetic image data generation via rendering techniques for training AI-based instance segmentation. Journal of Advanced Research in Applied Sciences and Engineering Technology, 62 (1). pp. 158-169. ISSN 2462-1943. (Published) https://doi.org/10.37934/araset.62.1.158169 10.37934/araset.62.1.158169 |
| spellingShingle | TJ Mechanical engineering and machinery TS Manufactures Kho, Dickson Yik Cheng Norazlianie, Sazali Ismayuzri, Ishak Saiful Anwar, Che Ghani Synthetic image data generation via rendering techniques for training AI-based instance segmentation |
| title | Synthetic image data generation via rendering techniques for training AI-based instance segmentation |
| title_full | Synthetic image data generation via rendering techniques for training AI-based instance segmentation |
| title_fullStr | Synthetic image data generation via rendering techniques for training AI-based instance segmentation |
| title_full_unstemmed | Synthetic image data generation via rendering techniques for training AI-based instance segmentation |
| title_short | Synthetic image data generation via rendering techniques for training AI-based instance segmentation |
| title_sort | synthetic image data generation via rendering techniques for training ai-based instance segmentation |
| topic | TJ Mechanical engineering and machinery TS Manufactures |
| url | http://umpir.ump.edu.my/id/eprint/42815/ http://umpir.ump.edu.my/id/eprint/42815/ http://umpir.ump.edu.my/id/eprint/42815/ http://umpir.ump.edu.my/id/eprint/42815/1/Synthetic%20image%20data%20generation%20via%20rendering%20techniques%20for%20training%20AI-based.pdf |