Degraded Image Semantic Segmentation With Dense-Gram Networks

@article{Guo2020DegradedIS,
  title={Degraded Image Semantic Segmentation With Dense-Gram Networks},
  author={Dazhou Guo and Yanting Pei and Kang Zheng and Hongkai Yu and Yuhang Lu and Song Wang},
  journal={IEEE Transactions on Image Processing},
  year={2020},
  volume={29},
  pages={782-795}
}
Degraded image semantic segmentation is of great importance in autonomous driving, highway navigation systems, and many other safety-related applications and it was not systematically studied before. In general, image degradations increase the difficulty of semantic segmentation, usually leading to decreased semantic segmentation accuracy. Therefore, performance on the underlying clean images can be treated as an upper bound of degraded image semantic segmentation. While the use of supervised… 
Double Similarity Distillation for Semantic Image Segmentation
TLDR
A simple yet general and effective knowledge distillation framework called double similarity distillation (DSD) to improve the classification accuracy of all existing compact networks by capturing the similarity knowledge in pixel and category dimensions, respectively.
Exposing Semantic Segmentation Failures via Maximum Discrepancy Competition
TLDR
This paper exposes failures of existing semantic segmentation methods in the open visual world under the constraint of very limited human labeling effort, and conducts a thorough MAD diagnosis of ten PASCAL VOC semantic segmentsation algorithms.
Multi-Type Self-Attention Guided Degraded Saliency Detection
TLDR
A new multi-type self-attention network, namely MSANet, is proposed for degraded saliency detection, which converts low-quality inputs to high-quality saliency maps directly in an end-to-end fashion.
Semantic Image Segmentation With Propagating Deep Aggregation
TLDR
A new architecture of feature aggregation is proposed, which is designed to deal with the problem that the information of each convolutional layer cannot be used reasonably and the shallow layer information is lost in the process of transmission.
CycleSegNet: Object Co-Segmentation With Cycle Refinement and Region Correspondence
Image co-segmentation is an active computer vision task that aims to segment the common objects from a set of images. Recently, researchers design various learning-based algorithms to undertake the
Effects of Image Size on Deep Learning
TLDR
The effects of late gadolinium enhancement (LGE)-magnetic resonance imaging (MRI) image size on deep learning -based fully automated quantification of myocardial infarction (MI) results are presented and a novel strategy is introduced and applied to remove extra class labels in interpolated ground truth (GT) segmentation masks.
DenseLightNet: A Light-Weight Vehicle Detection Network for Autonomous Driving
TLDR
Three new guidelines to determine the valid range of the group number are proposed, and a lightweight detection network—DenseLightNet—is designed based on these new design criteria.
Multiscale matters for part segmentation of instruments in robotic surgery
TLDR
This work introduces an end-to-end recurrent model that comprises a multiscale semantic segmentation network and a refinement model that uniformly transforms the input images in multiple scales into a semantic mask.
Learning to Detect Phone-related Pedestrian Distracted Behaviors with Synthetic Data
TLDR
A new deep learning model embedded with spatial-temporal feature learning and pose-aware transfer learning is proposed and could improve the state-of-the-art overall recognition accuracy from 84.27% to 96.67%.
...
...

References

SHOWING 1-10 OF 49 REFERENCES
A Review on Deep Learning Techniques Applied to Semantic Segmentation
TLDR
A review on deep learning methods for semantic segmentation applied to various application areas as well as mandatory background concepts to help researchers decide which are the ones that best suit their needs and their targets.
Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation
TLDR
This work extends DeepLabv3 by adding a simple yet effective decoder module to refine the segmentation results especially along object boundaries and applies the depthwise separable convolution to both Atrous Spatial Pyramid Pooling and decoder modules, resulting in a faster and stronger encoder-decoder network.
Context Encoding for Semantic Segmentation
TLDR
The proposed Context Encoding Module significantly improves semantic segmentation results with only marginal extra computation cost over FCN, and can improve the feature representation of relatively shallow networks for the image classification on CIFAR-10 dataset.
DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs
TLDR
This work addresses the task of semantic image segmentation with Deep Learning and proposes atrous spatial pyramid pooling (ASPP), which is proposed to robustly segment objects at multiple scales, and improves the localization of object boundaries by combining methods from DCNNs and probabilistic graphical models.
Examining the Impact of Blur on Recognition by Convolutional Networks
TLDR
It is found that by fine-tuning on a diverse mix of blurred images, convolutional neural networks can in fact learn to generate a blur invariant representation in their hidden layers.
RefineNet: Multi-path Refinement Networks for High-Resolution Semantic Segmentation
TLDR
RefineNet is presented, a generic multi-path refinement network that explicitly exploits all the information available along the down-sampling process to enable high-resolution prediction using long-range residual connections and introduces chained residual pooling, which captures rich background context in an efficient manner.
Small Object Sensitive Segmentation of Urban Street Scene With Spatial Adjacency Between Object Classes
TLDR
This paper proposes a new boundary-based metric that measures the level of spatial adjacency between each pair of object classes and finds that this metric is robust against object size-induced biases and develops a new method to enforce this metric into the segmentation loss.
Does Haze Removal Help CNN-Based Image Classification?
TLDR
From the experimental results, it is found that the existing image-dehazing methods cannot improve much the image-classification performance and sometimes even reduce the image’s performance.
Semantic Foggy Scene Understanding with Synthetic Data
TLDR
A complete pipeline to add synthetic fog to real, clear-weather images using incomplete depth information is developed and supervised learning with the authors' synthetic data significantly improves the performance of state-of-the-art CNN for SFSU on Foggy Driving.
SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation
TLDR
Quantitative assessments show that SegNet provides good performance with competitive inference time and most efficient inference memory-wise as compared to other architectures, including FCN and DeconvNet.
...
...