Monocular 3D depth for edge device with self-supervised learn by CNN and transformers with Lite-Mono

Monocular 3D depth for edge device with self-supervised learn by CNN and transformers with Lite-Mono

Lite-Mono: A Lightweight CNN and Transformer Architecture for Self-Supervised Monocular Depth Estimation
arXiv paper abstract https://arxiv.org/abs/2211.13202
arXiv PDF paper https://arxiv.org/pdf/2211.13202.pdf

Self-supervised monocular depth estimation that does not require ground-truth for training has attracted attention in recent years.

It is of high interest to design lightweight but effective models, so that they can be deployed on edge devices.

… In this paper … achieve comparable results with a lightweight architecture.

Specifically, … investigate the efficient combination of CNNs and Transformers, and design a hybrid architecture Lite-Mono.

… The former is used to extract rich multi-scale local features, and the latter takes advantage of the self-attention mechanism to encode long-range global information into the features.

… demonstrate that … full model outperforms Monodepth2 by a large margin in accuracy, with about 80% fewer trainable parameters.

Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website

LinkedIn https://www.linkedin.com/in/morris-lee-47877b7b

Photo by Tim Rüßmann on Unsplash