As a prerequisite for autonomous driving, scene understanding has attracted extensive research. With the rise of the convolutional neural network (CNN)-based deep learning technique, research on scene understanding has achieved significant progress. This paper aims to provide a comprehensive survey of deep learning-based approaches for scene understanding in autonomous driving. We categorize these works into four work streams, including object detection, full scene semantic segmentation, instance segmentation, and lane line segmentation. We discuss and analyze these works according to their characteristics, advantages and disadvantages, and basic frameworks. We also summarize the benchmark datasets and evaluation criteria used in the research community and make a performance comparison of some of the latest works. Lastly, we summarize the review work and provide a discussion on the future challenges of the research domain.
Visual odometry (VO) refers to incremental estimation of the motion state of an agent (e.g., vehicle and robot) by using image information, and is a key component of modern localization and navigation systems. Addressing the monocular VO problem, this paper presents a novel end-to-end network for estimation of camera ego-motion. The network learns the latent subspace of optical flow (OF) and models sequential dynamics so that the motion estimation is constrained by the relations between sequential images. We compute the OF field of consecutive images and extract the latent OF representation in a self-encoding manner. A Recurrent Neural Network is then followed to examine the OF changes, i.e., to conduct sequential learning. The extracted sequential OF subspace is used to compute the regression of the 6-dimensional pose vector. We derive three models with different network structures and different training schemes: LS-CNN-VO, LS-AE-VO, and LS-RCNN-VO. Particularly, we separately train the encoder in an unsupervised manner. By this means, we avoid non-convergence during the training of the whole network and allow more generalized and effective feature representation. Substantial experiments have been conducted on KITTI and Malaga datasets, and the results demonstrate that our LS-RCNN-VO outperforms the existing learning-based VO approaches.
MicroRNAs (miRs) have emerged as critical modulators of tumor initiation and progression in numerous types of human cancer, including clear cell renal cell carcinoma (ccRCC), which is the most common subtype of renal cell carcinoma. Cancerous inhibitor of protein phosphatase 2A (CIP2A) is a newly characterized oncoprotein and its overexpression has been reported to promote cellular epithelial-mesenchymal transition and the tumor progression of ccRCC. The present study examined the effects of miR-218 on CIP2A expression in ccRCC cells. The results demonstrated that the expression level of miR-218 was lower in ccRCC tissues compared with that in adjacent non-tumor renal tissues. In addition, it was identified that miR-128 could directly bind to the 3′-untranslated region of CIP2A. Furthermore, a negative correlation between the expression levels of miR-218 and CIP2A was detected in ccRCC. Additionally, the downregulation of CIP2A or overexpression of miR-218 in ccRCC cells was revealed to inhibit cell proliferation and migration. In summary, these data suggest that miR-218 serves a role in the regulation of CIP2A and elucidate its consequences on tumor progression, tumor cell proliferation and migration. These results indicate that miR-218 may exhibit potential as a molecular target for the treatment of ccRCC.
While MRI contrast agents such as those based on Gadolinium are needed to enhance the detection of structural and functional brain lesions, there are rising concerns over their safety. Here, we hypothesize that a deep learning model, trained using quantitative steady-state contrast-enhanced MRI datasets in mice and humans, could generate contrast-equivalent information from a single non-contrast MRI scan. The model was first trained, optimized, and validated in mice. It was then transferred and adapted to human data, and we find that it can substitute Gadolinium-based contrast agents for detecting functional lesions caused by aging, Schizophrenia, or Alzheimer’s disease, and, for enhancing structural lesions caused by brain or breast tumors. Since derived from a commonly-acquired MRI, this framework has the potential for broad clinical utility and can be applied retrospectively to research scans across a host of diseases.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.