Deep Neighbor Layer Aggregation for Lightweight Self-Supervised Monocular Depth Estimation

TL;DR

This paper introduces a lightweight convolutional network with contextual feature fusion and channel attention for efficient self-supervised monocular depth estimation, outperforming larger models on KITTI with fewer parameters.

Contribution

The proposed method combines high- and low-resolution features with lightweight attention to reduce parameters while maintaining accuracy in depth estimation.

Findings

Outperforms larger models like Monodepth2 on KITTI

Uses only 30 parameters, significantly fewer than competitors

Achieves better depth estimation results with efficient architecture

Abstract

With the frequent use of self-supervised monocular depth estimation in robotics and autonomous driving, the model's efficiency is becoming increasingly important. Most current approaches apply much larger and more complex networks to improve the precision of depth estimation. Some researchers incorporated Transformer into self-supervised monocular depth estimation to achieve better performance. However, this method leads to high parameters and high computation. We present a fully convolutional depth estimation network using contextual feature fusion. Compared to UNet++ and HRNet, we use high-resolution and low-resolution features to reserve information on small targets and fast-moving objects instead of long-range fusion. We further promote depth estimation results employing lightweight channel attention based on convolution in the decoder stage. Our method reduces the parameters without sacrificing accuracy. Experiments on the KITTI benchmark show that our method can get better results than many large models, such as Monodepth2, with only 30 parameters. The source code is available at https://github.com/boyagesmile/DNA-Depth.