PanDepth: Joint Panoptic Segmentation and Depth Completion
This work addresses the need for efficient multi-task perception in autonomous driving, though it appears incremental as it combines existing tasks into a single model.
The authors tackled the problem of holistic 3D scene understanding in autonomous driving by proposing a multi-task model that jointly performs panoptic segmentation and depth completion from RGB images and sparse depth maps, achieving high accuracy without significantly increasing computational cost on the Virtual KITTI 2 dataset.
Understanding 3D environments semantically is pivotal in autonomous driving applications where multiple computer vision tasks are involved. Multi-task models provide different types of outputs for a given scene, yielding a more holistic representation while keeping the computational cost low. We propose a multi-task model for panoptic segmentation and depth completion using RGB images and sparse depth maps. Our model successfully predicts fully dense depth maps and performs semantic segmentation, instance segmentation, and panoptic segmentation for every input frame. Extensive experiments were done on the Virtual KITTI 2 dataset and we demonstrate that our model solves multiple tasks, without a significant increase in computational cost, while keeping high accuracy performance. Code is available at https://github.com/juanb09111/PanDepth.git