Learning Inverse Laplacian Pyramid for Progressive Depth Completion

TL;DR

LP-Net introduces a multi-scale, progressive depth completion framework using Laplacian Pyramid decomposition, achieving state-of-the-art results with improved efficiency by globally capturing scene context and refining details iteratively.

Contribution

The paper presents LP-Net, a novel multi-scale depth completion method with two new modules, outperforming existing approaches in accuracy and computational efficiency.

Findings

Achieves SOTA performance on KITTI, NYUv2, and TOFDC benchmarks.

Ranks 1st among peer-reviewed methods on KITTI leaderboard.

Demonstrates superior efficiency compared to propagation-based methods.

Abstract

Depth completion endeavors to reconstruct a dense depth map from sparse depth measurements, leveraging the information provided by a corresponding color image. Existing approaches mostly hinge on single-scale propagation strategies that iteratively ameliorate initial coarse depth estimates through pixel-level message passing. Despite their commendable outcomes, these techniques are frequently hampered by computational inefficiencies and a limited grasp of scene context. To circumvent these challenges, we introduce LP-Net, an innovative framework that implements a multi-scale, progressive prediction paradigm based on Laplacian Pyramid decomposition. Diverging from propagation-based approaches, LP-Net initiates with a rudimentary, low-resolution depth prediction to encapsulate the global scene context, subsequently refining this through successive upsampling and the reinstatement of high-frequency details at incremental scales. We have developed two novel modules to bolster this strategy: 1) the Multi-path Feature Pyramid module, which segregates feature maps into discrete pathways, employing multi-scale transformations to amalgamate comprehensive spatial information, and 2) the Selective Depth Filtering module, which dynamically learns to apply both smoothness and sharpness filters to judiciously mitigate noise while accentuating intricate details. By integrating these advancements, LP-Net not only secures state-of-the-art (SOTA) performance across both outdoor and indoor benchmarks such as KITTI, NYUv2, and TOFDC, but also demonstrates superior computational efficiency. At the time of submission, LP-Net ranks 1st among all peer-reviewed methods on the official KITTI leaderboard.