Transfer Learning from Simulated to Real Scenes for Monocular 3D Object Detection
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910580852391936 |
|---|---|
| author | Mohamed, Sondos Zimmer, Walter Greer, Ross Ghita, Ahmed Alaaeldin Castrillón-Santana, Modesto Trivedi, Mohan Knoll, Alois Carta, Salvatore Mario Marras, Mirko |
| author_facet | Mohamed, Sondos Zimmer, Walter Greer, Ross Ghita, Ahmed Alaaeldin Castrillón-Santana, Modesto Trivedi, Mohan Knoll, Alois Carta, Salvatore Mario Marras, Mirko |
| contents | Accurately detecting 3D objects from monocular images in dynamic roadside scenarios remains a challenging problem due to varying camera perspectives and unpredictable scene conditions. This paper introduces a two-stage training strategy to address these challenges. Our approach initially trains a model on the large-scale synthetic dataset, RoadSense3D, which offers a diverse range of scenarios for robust feature learning. Subsequently, we fine-tune the model on a combination of real-world datasets to enhance its adaptability to practical conditions. Experimental results of the Cube R-CNN model on challenging public benchmarks show a remarkable improvement in detection performance, with a mean average precision rising from 0.26 to 12.76 on the TUM Traffic A9 Highway dataset and from 2.09 to 6.60 on the DAIR-V2X-I dataset when performing transfer learning. Code, data, and qualitative video results are available on the project website: https://roadsense3d.github.io. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_15637 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Transfer Learning from Simulated to Real Scenes for Monocular 3D Object Detection Mohamed, Sondos Zimmer, Walter Greer, Ross Ghita, Ahmed Alaaeldin Castrillón-Santana, Modesto Trivedi, Mohan Knoll, Alois Carta, Salvatore Mario Marras, Mirko Computer Vision and Pattern Recognition Accurately detecting 3D objects from monocular images in dynamic roadside scenarios remains a challenging problem due to varying camera perspectives and unpredictable scene conditions. This paper introduces a two-stage training strategy to address these challenges. Our approach initially trains a model on the large-scale synthetic dataset, RoadSense3D, which offers a diverse range of scenarios for robust feature learning. Subsequently, we fine-tune the model on a combination of real-world datasets to enhance its adaptability to practical conditions. Experimental results of the Cube R-CNN model on challenging public benchmarks show a remarkable improvement in detection performance, with a mean average precision rising from 0.26 to 12.76 on the TUM Traffic A9 Highway dataset and from 2.09 to 6.60 on the DAIR-V2X-I dataset when performing transfer learning. Code, data, and qualitative video results are available on the project website: https://roadsense3d.github.io. |
| title | Transfer Learning from Simulated to Real Scenes for Monocular 3D Object Detection |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2408.15637 |