PanDepth: Joint Panoptic Segmentation and Depth Completion

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Lagos, Juan, Rahtu, Esa
Natura: Preprint
Pubblicazione: 2022
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911994826719232
author Lagos, Juan
Rahtu, Esa
author_facet Lagos, Juan
Rahtu, Esa
contents 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
format Preprint
id arxiv_https___arxiv_org_abs_2212_14180
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle PanDepth: Joint Panoptic Segmentation and Depth Completion
Lagos, Juan
Rahtu, Esa
Computer Vision and Pattern Recognition
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
title PanDepth: Joint Panoptic Segmentation and Depth Completion
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2212.14180