Search

There is great interest in automatically detecting road weather and understanding its impacts on the overall safety of the transport network. This can, for example, support road condition-based maintenance or even serve as detection systems that assist safe driving during adverse climate conditions. In computer vision, previous work has demonstrated the effectiveness of deep learning in predicting weather conditions from outdoor images. However, training deep learning models to accurately predict weather conditions using real-world road-facing images is difficult due to: (1) the simultaneous occurrence of multiple weather conditions; (2) imbalanced occurrence of weather conditions throughout the year; and (3) road idiosyncrasies, such as road layouts, illumination, and road objects, etc. In this paper, we explore the use of a focal loss function to force the learning process to focus on weather instances that are hard to learn with the objective of helping address data imbalances. In addition, we explore the attention mechanism for pixel-based dynamic weight adjustment to handle road idiosyncrasies using state-of-the-art vision transformer models. Experiments with a novel multi-label road weather dataset show that focal loss significantly increases the accuracy of computer vision approaches for imbalanced weather conditions. Furthermore, vision transformers outperform current state-of-the-art convolutional neural networks in predicting weather conditions with a validation accuracy of 92% and an F1-score of 81.22%, which is impressive considering the imbalanced nature of the dataset. [ABSTRACT FROM AUTHOR]

A vital and challenging task in computer vision is 3D Object Classification and Retrieval, with many practical applications such as an intelligent robot, autonomous driving, multimedia contents processing and retrieval, and augmented/mixed reality. Various deep learning methods were introduced for solving classification and retrieval problems of 3D objects. Almost all view-based methods use many views to handle spatial loss, although they perform the best among current techniques such as View-based, Voxelization, and Point Cloud methods. Many views make network structure more complicated due to the parallel Convolutional Neural Network (CNN). We propose a novel method that combines a Global Point Signature Plus with a Deep Wide Residual Network, namely GPSP-DWRN, in this paper. Global Point Signature Plus (GPSPlus) is a novel descriptor because it can capture more shape information of the 3D object for a single view. First, an original 3D model was converted into a colored one by applying GPSPlus. Then, a 32 × 32 × 3 matrix stored the obtained 2D projection of this color 3D model. This matrix was the input data of a Deep Residual Network, which used a single CNN structure. We evaluated the GPSP-DWRN for a retrieval task using the Shapnetcore55 dataset, while using two well-known datasets—ModelNet10 and ModelNet40 for a classification task. Based on our experimental results, our framework performed better than the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Semantic segmentation, which refers to pixel-wise classification of an image, is a fundamental topic in computer vision owing to its growing importance in the robot vision and autonomous driving sectors. It provides rich information about objects in the scene such as object boundary, category, and location. Recent methods for semantic segmentation often employ an encoder-decoder structure using deep convolutional neural networks. The encoder part extracts features of the image using several filters and pooling operations, whereas the decoder part gradually recovers the low-resolution feature maps of the encoder into a full input resolution feature map for pixel-wise prediction. However, the encoder-decoder variants for semantic segmentation suffer from severe spatial information loss, caused by pooling operations or stepwise convolutions, and does not consider the context in the scene. In this paper, we propose a novel dense upsampling convolution method based on a guided filter to effectively preserve the spatial information of the image in the network. We further propose a novel local context convolution method that not only covers larger-scale objects in the scene but covers them densely for precise object boundary delineation. Theoretical analyses and experimental results on several benchmark datasets verify the effectiveness of our method. Qualitatively, our approach delineates object boundaries at a level of accuracy that is beyond the current excellent methods. Quantitatively, we report a new record of 82.86% and 81.62% of pixel accuracy on ADE20K and Pascal-Context benchmark datasets, respectively. In comparison with the state-of-the-art methods, the proposed method offers promising improvements. [ABSTRACT FROM AUTHOR]