Confidence Guided Depth Completion Network

TL;DR

This paper introduces a two-stage, confidence-guided depth completion network that produces accurate dense depth maps efficiently, outperforming top models in speed while maintaining competitive accuracy.

Contribution

It presents a novel two-stage depth completion network that leverages confidence maps for refinement, achieving faster computation with competitive accuracy.

Findings

Faster depth completion compared to top models

Competitive accuracy on KITTI benchmark

Effective confidence-guided refinement process

Abstract

The paper proposes an image-guided depth completion method to estimate accurate dense depth maps with fast computation time. The proposed network has two-stage structure. The first stage predicts a first depth map. Then, the second stage further refines the first depth map using the confidence maps. The second stage consists of two layers, each of which focuses on different regions and generates a refined depth map and a confidence map. The final depth map is obtained by combining two depth maps from the second stage using the corresponding confidence maps. Compared with the top-ranked models on the KITTI depth completion online leaderboard, the proposed model shows much faster computation time and competitive performance.