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.
@article{arxiv.2408.15637,
title = {Transfer Learning from Simulated to Real Scenes for Monocular 3D Object Detection},
author = {Sondos Mohamed and Walter Zimmer and Ross Greer and Ahmed Alaaeldin Ghita and Modesto Castrillón-Santana and Mohan Trivedi and Alois Knoll and Salvatore Mario Carta and Mirko Marras},
journal= {arXiv preprint arXiv:2408.15637},
year = {2024}
}
Comments
18 pages. Accepted for ECVA European Conference on Computer Vision 2024 (ECCV'24)