Cost-effective annotation of fisheye images for object detection
Nowadays, fisheye image has become commonly used in the 3D reality capturing field. Although AI integration for image recognition has become mature with normal images, providing available annotated dataset and pre-trained models, its application for fisheye images is rarely seen. While the object de...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2024-12-01
|
| Series: | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| Online Access: | https://isprs-archives.copernicus.org/articles/XLVIII-2-W8-2024/491/2024/isprs-archives-XLVIII-2-W8-2024-491-2024.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850256438770794496 |
|---|---|
| author | K. Zhang A. Elalailyi L. Perfetti F. Fassi |
| author_facet | K. Zhang A. Elalailyi L. Perfetti F. Fassi |
| author_sort | K. Zhang |
| collection | DOAJ |
| description | Nowadays, fisheye image has become commonly used in the 3D reality capturing field. Although AI integration for image recognition has become mature with normal images, providing available annotated dataset and pre-trained models, its application for fisheye images is rarely seen. While the object detection models have generalization ability, dealing with barrel distortion requires specific data for fine-tuning. This paper seeks to acquire prior knowledge from normal image and transfer it to the application that deal with fisheye images. This research is devoted to test the annotation shape that could possibly improve the accuracy when representing the shape of objects. It also seeks a way to prove that the annotation can be converted to fisheye images, resulted into a pre-process, which will facilitate the data preparation process. The tests involve annotations with standard box and quadrilateral polygon, the later turned out to be preserving most of the wanted image content after the conversion. The test result shows that the model trained on converted annotations using quadrilateral polygons, compared to detection model trained on non-converted ones, improves the mean average precision by 8%. |
| format | Article |
| id | doaj-art-80e4cd57485f4549997c8668ce00d9a2 |
| institution | OA Journals |
| issn | 1682-1750 2194-9034 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Copernicus Publications |
| record_format | Article |
| series | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| spelling | doaj-art-80e4cd57485f4549997c8668ce00d9a22025-08-20T01:56:38ZengCopernicus PublicationsThe International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences1682-17502194-90342024-12-01XLVIII-2-W8-202449149810.5194/isprs-archives-XLVIII-2-W8-2024-491-2024Cost-effective annotation of fisheye images for object detectionK. Zhang0A. Elalailyi1L. Perfetti2F. Fassi33D Survey Group, ABC Department, Politecnico di Milano, Via Ponzio 31, 20133 Milano, Italy3D Survey Group, ABC Department, Politecnico di Milano, Via Ponzio 31, 20133 Milano, ItalyDepartment of Civil, Architectural, Environmental Engineering and Mathematics (DICATAM), Università degli Studi di Brescia, 25123 Brescia, Italy3D Survey Group, ABC Department, Politecnico di Milano, Via Ponzio 31, 20133 Milano, ItalyNowadays, fisheye image has become commonly used in the 3D reality capturing field. Although AI integration for image recognition has become mature with normal images, providing available annotated dataset and pre-trained models, its application for fisheye images is rarely seen. While the object detection models have generalization ability, dealing with barrel distortion requires specific data for fine-tuning. This paper seeks to acquire prior knowledge from normal image and transfer it to the application that deal with fisheye images. This research is devoted to test the annotation shape that could possibly improve the accuracy when representing the shape of objects. It also seeks a way to prove that the annotation can be converted to fisheye images, resulted into a pre-process, which will facilitate the data preparation process. The tests involve annotations with standard box and quadrilateral polygon, the later turned out to be preserving most of the wanted image content after the conversion. The test result shows that the model trained on converted annotations using quadrilateral polygons, compared to detection model trained on non-converted ones, improves the mean average precision by 8%.https://isprs-archives.copernicus.org/articles/XLVIII-2-W8-2024/491/2024/isprs-archives-XLVIII-2-W8-2024-491-2024.pdf |
| spellingShingle | K. Zhang A. Elalailyi L. Perfetti F. Fassi Cost-effective annotation of fisheye images for object detection The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| title | Cost-effective annotation of fisheye images for object detection |
| title_full | Cost-effective annotation of fisheye images for object detection |
| title_fullStr | Cost-effective annotation of fisheye images for object detection |
| title_full_unstemmed | Cost-effective annotation of fisheye images for object detection |
| title_short | Cost-effective annotation of fisheye images for object detection |
| title_sort | cost effective annotation of fisheye images for object detection |
| url | https://isprs-archives.copernicus.org/articles/XLVIII-2-W8-2024/491/2024/isprs-archives-XLVIII-2-W8-2024-491-2024.pdf |
| work_keys_str_mv | AT kzhang costeffectiveannotationoffisheyeimagesforobjectdetection AT aelalailyi costeffectiveannotationoffisheyeimagesforobjectdetection AT lperfetti costeffectiveannotationoffisheyeimagesforobjectdetection AT ffassi costeffectiveannotationoffisheyeimagesforobjectdetection |