Published on Sun Mar 21 2021

Deep Dense Multi-scale Network for Snow Removal Using Semantic and Geometric Priors

Kaihao Zhang, Rongqing Li, Yanjiang Yu, Wenhan Luo, Changsheng Li, Hongdong Li
0
0
0
Abstract

Images captured in snowy days suffer from noticeable degradation of scene visibility, which degenerates the performance of current vision-based intelligent systems. Removing snow from images thus is an important topic in computer vision. In this paper, we propose a Deep Dense Multi-Scale Network (\textbf{DDMSNet}) for snow removal by exploiting semantic and geometric priors. As images captured in outdoor often share similar scenes and their visibility varies with depth from camera, such semantic and geometric information provides a strong prior for snowy image restoration. We incorporate the semantic and geometric maps as input and learn the semantic-aware and geometry-aware representation to remove snow. In particular, we first create a coarse network to remove snow from the input images. Then, the coarsely desnowed images are fed into another network to obtain the semantic and geometric labels. Finally, we design a DDMSNet to learn semantic-aware and geometry-aware representation via a self-attention mechanism to produce the final clean images. Experiments evaluated on public synthetic and real-world snowy images verify the superiority of the proposed method, offering better results both quantitatively and qualitatively.

Tue Aug 15 2017
Computer Vision
DesnowNet: Context-Aware Deep Network for Snow Removal
Hand-crafted features are still the mainstream for snow removal. In response, we have designed a multistage network codenamed DesnowNet. We also differentiate snow particles into attributes of translucency and chromatic aberration for accurate estimation.
0
0
0
Sun Feb 28 2021
Computer Vision
Snowy Night-to-Day Translator and Semantic Segmentation Label Similarity for Snow Hazard Indicator
In 2021, Japan recorded more than three times as much snowfall as usual. At the night time zone, the temperature drops and the road surface tends to freeze. The poor visibility caused by the snow triggers traffic accidents.
0
0
0
Thu Jan 14 2021
Computer Vision
Road Surface Translation Under Snow-covered and Semantic Segmentation for Snow Hazard Index
Study proposes a deep learning application to automatically calculate a snow hazard ratio. First, the road surface hidden under snow is translated using agenerative adversarial network, pix2pix. Second, snow-covered and road surface classes are detected by semantic segmentation.
0
0
0
Fri Mar 26 2021
Computer Vision
Marine Snow Removal Benchmarking Dataset
0
0
0
Thu Jan 23 2020
Neural Networks
Cloud and Cloud Shadow Segmentation for Remote Sensing Imagery via Filtered Jaccard Loss Function and Parametric Augmentation
Current methods for cloud/shadow identification are not as accurate as they should, especially in the presence of snow and haze. Our method benefits from a convolutional neural network, Cloud-Net+ that is trained with a novel loss function.
0
0
0
Fri Sep 13 2019
Computer Vision
Video Rain/Snow Removal by Transformed Online Multiscale Convolutional Sparse Coding
Video rain/snow removal from surveillance videos is an important task in the computer vision community. Various methods have been investigated, but most only consider consistent background scenes. Rain/s snow captured from practical surveillance cameras is always dynamic in time.
0
0
0