Real-time single image depth perception in the wild with handheld devices

Aleotti, Filippo; Zaccaroni, Giulio; Bartolomei, Luca; Poggi, Matteo; Tosi, Fabio; Mattoccia, Stefano

doi:10.48550/arxiv.2006.05724

Cited by 2 publications

(2 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(Wofk et al 2019) proposed a lightweight supervised depth estimation network, who use encoder-decoder architecture and include 1.34M parameters after pruning. In unsupervised filed, (Poggi et al 2018;Aleotti et al 2020) proposed PyDNet with 1.9M parameters. Although the above two works have less parameters, their performance also decreases a lot.…”

Section: Lightweight Network For Depth Estimationmentioning

confidence: 99%

HR-Depth: High Resolution Self-Supervised Monocular Depth Estimation

Lyu

Liu

Wang

et al. 2021

AAAI

165

View full text Add to dashboard Cite

Self-supervised learning shows great potential in monocular depth estimation, using image sequences as the only source of supervision. Although people try to use the high-resolution image for depth estimation, the accuracy of prediction has not been significantly improved. In this work, we find the core reason comes from the inaccurate depth estimation in large gradient regions, making the bilinear interpolation error gradually disappear as the resolution increases. To obtain more accurate depth estimation in large gradient regions, it is necessary to obtain high-resolution features with spatial and semantic information. Therefore, we present an improved DepthNet, HR-Depth, with two effective strategies: (1) re-design the skip-connection in DepthNet to get better high-resolution features and (2) propose feature fusion Squeeze-and-Excitation(fSE) module to fuse feature more efficiently. Using Resnet-18 as the encoder, HR-Depth surpasses all previous state-of-the-art(SoTA) methods with the least parameters at both high and low resolution. Moreover, previous SoTA methods are based on fairly complex and deep networks with a mass of parameters which limits their real applications. Thus we also construct a lightweight network which uses MobileNetV3 as encoder. Experiments show that the lightweight network can perform on par with many large models like Monodepth2 at high-resolution with only20%parameters. All codes and models will be available at https://github.com/shawLyu/HR-Depth.

show abstract

Section: Lightweight Network For Depth Estimationmentioning

confidence: 99%

HR-Depth: High Resolution Self-Supervised Monocular Depth Estimation

Lyu

Liu

Wang

et al. 2021

AAAI

165

View full text Add to dashboard Cite

show abstract

Section: Lightweight Network For Depth Estimationmentioning

confidence: 99%

HR-Depth: High Resolution Self-Supervised Monocular Depth Estimation

Lyu

Liu

Wang

et al. 2020

Preprint

View full text Add to dashboard Cite

Self-supervised learning shows great potential in monocular depth estimation, using image sequences as the only source of supervision. Although people try to use the high-resolution image for depth estimation, the accuracy of prediction has not been significantly improved. In this work, we find the core reason comes from the inaccurate depth estimation in large gradient regions, making the bilinear interpolation error gradually disappear as the resolution increases. To obtain more accurate depth estimation in large gradient regions, it is necessary to obtain high-resolution features with spatial and semantic information. Therefore, we present an improved DepthNet, HR-Depth, with two effective strategies: (1) redesign the skip-connection in DepthNet to get better highresolution features and (2) propose feature fusion Squeezeand-Excitation(fSE) module to fuse feature more efficiently. Using Resnet-18 as the encoder, HR-Depth surpasses all previous state-of-the-art(SoTA) methods with the least parameters at both high and low resolution. Moreover, previous state-of-the-art methods are based on fairly complex and deep networks with a mass of parameters which limits their real applications. Thus we also construct a lightweight network which uses MobileNetV3 as encoder. Experiments show that the lightweight network can perform on par with many large models like Monodepth2 at high-resolution with only 20% parameters. All codes and models will be available at https: //github.com/shawLyu/HR-Depth.

show abstract

Real-time single image depth perception in the wild with handheld devices

Cited by 2 publications

References 38 publications

HR-Depth: High Resolution Self-Supervised Monocular Depth Estimation

HR-Depth: High Resolution Self-Supervised Monocular Depth Estimation

HR-Depth: High Resolution Self-Supervised Monocular Depth Estimation

Contact Info

Product

Resources

About