/RGBDVS-fusion

Implementation code for: Semantic Segmentation and Depth Estimation with RGB and DVS Sensor Fusion for Multi-view Driving Perception, Proc. Asian Conf. Pattern Recognition (ACPR), 2021.

Primary LanguagePythonMIT LicenseMIT

Semantic Segmentation and Depth Estimation with RGB and DVS Sensor Fusion for Multi-view Driving Perception

O. Natan and J. Miura, "Semantic Segmentation and Depth Estimation with RGB and DVS Sensor Fusion for Multi-view Driving Perception," in Proc. Asian Conf. Pattern Recognition (ACPR), Jeju Island, South Korea, Nov. 2021, pp. 352–365. [paper]

Related works:

  1. O. Natan and J. Miura, “Towards Compact Autonomous Driving Perception with Balanced Learning and Multi-sensor Fusion,” IEEE Trans. Intelligent Transportation Systems, 2022. [paper] [code]
  2. O. Natan and J. Miura, “End-to-end Autonomous Driving with Semantic Depth Cloud Mapping and Multi-agent,” IEEE Trans. Intelligent Vehicles, 2022. [paper] [code]
  3. O. Natan and J. Miura, “DeepIPC: Deeply Integrated Perception and Control for Mobile Robot in Real Environments,” arXiv:2207.09934, 2022. [paper]

Steps:

  1. Download the dataset and extract to subfolder dataset
  2. Open train.py and check the configuration setting inside
  3. python3 train.py, model and other metadata will be saved in subfolder model
  4. python3 predict.py, prediction result will be saved in subfolder prediction