A synthetic digital city dataset for robustness and generalisation of depth estimation models
Abstract Existing monocular depth estimation driving datasets are limited in the number of images and the diversity of driving conditions. The images of datasets are commonly in a low resolution and the depth maps are sparse. To overcome these limitations, we produce a Synthetic Digital City Dataset...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-03-01
|
| Series: | Scientific Data |
| Online Access: | https://doi.org/10.1038/s41597-024-03025-5 |
| _version_ | 1797259531495931904 |
|---|---|
| author | Jihao Li Jincheng Hu Yanjun Huang Zheng Chen Bingzhao Gao Jingjing Jiang Yuanjian Zhang |
| author_facet | Jihao Li Jincheng Hu Yanjun Huang Zheng Chen Bingzhao Gao Jingjing Jiang Yuanjian Zhang |
| author_sort | Jihao Li |
| collection | DOAJ |
| description | Abstract Existing monocular depth estimation driving datasets are limited in the number of images and the diversity of driving conditions. The images of datasets are commonly in a low resolution and the depth maps are sparse. To overcome these limitations, we produce a Synthetic Digital City Dataset (SDCD) which was collected under 6 different weather driving conditions, and 6 common adverse perturbations caused by the data transmission. SDCD provides a total of 930 K high-resolution RGB images and corresponding perfect observed depth maps. The evaluation shows that depth estimation models which are trained on SDCD provide a clearer, smoother, and more precise long-range depth estimation compared to those trained on one of the best-known driving datasets KITTI. Moreover, we provide a benchmark to investigate the performance of depth estimation models in different adverse driving conditions. Instead of collecting data from the real world, we generate the SDCD under severe driving conditions with perfect observed data in the digital world, enhancing depth estimation for autonomous driving. |
| first_indexed | 2024-04-24T23:10:55Z |
| format | Article |
| id | doaj.art-d6810f68c2d7430cbf192dcbd57759e7 |
| institution | Directory Open Access Journal |
| issn | 2052-4463 |
| language | English |
| last_indexed | 2024-04-24T23:10:55Z |
| publishDate | 2024-03-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Data |
| spelling | doaj.art-d6810f68c2d7430cbf192dcbd57759e72024-03-17T12:14:14ZengNature PortfolioScientific Data2052-44632024-03-0111111310.1038/s41597-024-03025-5A synthetic digital city dataset for robustness and generalisation of depth estimation modelsJihao Li0Jincheng Hu1Yanjun Huang2Zheng Chen3Bingzhao Gao4Jingjing Jiang5Yuanjian Zhang6Department of Aeronautical and Automotive Engineering, Loughborough UniversityDepartment of Aeronautical and Automotive Engineering, Loughborough UniversitySchool of Automotive Studies, Tongji UniversityFaculty of Transportation Engineering, Kunming University of Science and TechnologySchool of Automotive Studies, Tongji UniversityDepartment of Aeronautical and Automotive Engineering, Loughborough UniversityDepartment of Aeronautical and Automotive Engineering, Loughborough UniversityAbstract Existing monocular depth estimation driving datasets are limited in the number of images and the diversity of driving conditions. The images of datasets are commonly in a low resolution and the depth maps are sparse. To overcome these limitations, we produce a Synthetic Digital City Dataset (SDCD) which was collected under 6 different weather driving conditions, and 6 common adverse perturbations caused by the data transmission. SDCD provides a total of 930 K high-resolution RGB images and corresponding perfect observed depth maps. The evaluation shows that depth estimation models which are trained on SDCD provide a clearer, smoother, and more precise long-range depth estimation compared to those trained on one of the best-known driving datasets KITTI. Moreover, we provide a benchmark to investigate the performance of depth estimation models in different adverse driving conditions. Instead of collecting data from the real world, we generate the SDCD under severe driving conditions with perfect observed data in the digital world, enhancing depth estimation for autonomous driving.https://doi.org/10.1038/s41597-024-03025-5 |
| spellingShingle | Jihao Li Jincheng Hu Yanjun Huang Zheng Chen Bingzhao Gao Jingjing Jiang Yuanjian Zhang A synthetic digital city dataset for robustness and generalisation of depth estimation models Scientific Data |
| title | A synthetic digital city dataset for robustness and generalisation of depth estimation models |
| title_full | A synthetic digital city dataset for robustness and generalisation of depth estimation models |
| title_fullStr | A synthetic digital city dataset for robustness and generalisation of depth estimation models |
| title_full_unstemmed | A synthetic digital city dataset for robustness and generalisation of depth estimation models |
| title_short | A synthetic digital city dataset for robustness and generalisation of depth estimation models |
| title_sort | synthetic digital city dataset for robustness and generalisation of depth estimation models |
| url | https://doi.org/10.1038/s41597-024-03025-5 |
| work_keys_str_mv | AT jihaoli asyntheticdigitalcitydatasetforrobustnessandgeneralisationofdepthestimationmodels AT jinchenghu asyntheticdigitalcitydatasetforrobustnessandgeneralisationofdepthestimationmodels AT yanjunhuang asyntheticdigitalcitydatasetforrobustnessandgeneralisationofdepthestimationmodels AT zhengchen asyntheticdigitalcitydatasetforrobustnessandgeneralisationofdepthestimationmodels AT bingzhaogao asyntheticdigitalcitydatasetforrobustnessandgeneralisationofdepthestimationmodels AT jingjingjiang asyntheticdigitalcitydatasetforrobustnessandgeneralisationofdepthestimationmodels AT yuanjianzhang asyntheticdigitalcitydatasetforrobustnessandgeneralisationofdepthestimationmodels AT jihaoli syntheticdigitalcitydatasetforrobustnessandgeneralisationofdepthestimationmodels AT jinchenghu syntheticdigitalcitydatasetforrobustnessandgeneralisationofdepthestimationmodels AT yanjunhuang syntheticdigitalcitydatasetforrobustnessandgeneralisationofdepthestimationmodels AT zhengchen syntheticdigitalcitydatasetforrobustnessandgeneralisationofdepthestimationmodels AT bingzhaogao syntheticdigitalcitydatasetforrobustnessandgeneralisationofdepthestimationmodels AT jingjingjiang syntheticdigitalcitydatasetforrobustnessandgeneralisationofdepthestimationmodels AT yuanjianzhang syntheticdigitalcitydatasetforrobustnessandgeneralisationofdepthestimationmodels |