Digging Into Uncertainty-Based Pseudo-Label for Robust Stereo Matching.

Due to the domain differences and unbalanced disparity distribution across multiple datasets, current stereo matching approaches are commonly limited to a specific dataset and generalize poorly to others. Such domain shift issue is usually addressed by substantial adaptation on costly target-domain ground-truth data, which cannot be easily obtained in practical settings. In this paper, we propose to dig into uncertainty estimation for robust stereo matching. Specifically, to balance the disparity distribution, we employ a pixel-level uncertainty estimation to adaptively adjust the next stage disparity searching space, in this way driving the network progressively prune out the space of unlikely correspondences. Then, to solve the limited ground truth data, an uncertainty-based pseudo-label is proposed to adapt the pre-trained model to the new domain, where pixel-level and area-level uncertainty estimation are proposed to filter out the high-uncertainty pixels of predicted disparity maps and generate sparse while reliable pseudo-labels to align the domain gap. Experimentally, our method shows strong cross-domain, adapt, and joint generalization and obtains 1st place on the stereo task of Robust Vision Challenge 2020. Additionally, our uncertainty-based pseudo-labels can be extended to train monocular depth estimation networks in an unsupervised way and even achieves comparable performance with the supervised methods.

TUSR-Net: Triple Unfolding Single Image Dehazing with Self-Regularization and Dual Feature to Pixel Attention.

Single image dehazing is a challenging and illposed problem due to severe information degeneration of images captured in hazy conditions. Remarkable progresses have been achieved by deep-learning based image dehazing methods, where residual learning is commonly used to separate the hazy image into clear and haze components. However, the nature of low similarity between haze and clear components is commonly neglected, while the lack of constraint of contrastive peculiarity between the two components always restricts the performance of these approaches. To deal with these problems, we propose an end-to-end self-regularized network (TUSR-Net) which exploits the contrastive peculiarity of different components of the hazy image, i.e, self-regularization (SR). In specific, the hazy image is separated into clear and hazy components and constraint between different image components, i.e., self-regularization, is leveraged to pull the recovered clear image closer to groundtruth, which largely promotes the performance of image dehazing. Meanwhile, an effective triple unfolding framework combined with dual feature to pixel attention is proposed to intensify and fuse the intermediate information in feature, channel and pixel levels, respectively, thus features with better representational ability can be obtained. Our TUSR-Net achieves better trade-off between performance and parameter size with weight-sharing strategy and is much more flexible. Experiments on various benchmarking datasets demonstrate the superiority of our TUSR-Net over state-of-the-art single image dehazing methods.

Rethinking Training Strategy in Stereo Matching.

In stereo matching, various learning-based approaches have shown impressive performance in solving traditional difficulties on multiple datasets. While most progress is obtained on a specific dataset with a dataset-specific network design, the performance on the single dataset and cross dataset affected by training strategy is often ignored. In this article, we analyze the relationship between different training strategies and performance by retraining some representative state-of-the-art methods (e.g., geometry and context network (GC-Net), pyramid stereo matching network (PSM-Net), and guided aggregation network (GA-Net), etc.). According to our research, it is surprising that the performance of networks on single or cross datasets is significantly improved by pre-training and data augmentation without any particular structure acquirement. Based on this discovery, we improve our previous non-local context attention network (NLCA-Net) to NLCA-Net v2 and train it with the novel strategy and rethink the training strategy of stereo matching concurrently. The quantitative experiments demonstrate that: 1) our model is capable of reaching top performance on both the single dataset and the multiple datasets with the same parameters in this study, which also won the 2nd place in the stereo task of the ECCV Robust vision Challenge 2020 (RVC 2020); and 2) on small datasets (e.g., KITTI, ETH3D, and Middlebury), the model's generalization and robustness are significantly affected by pre-training and data augmentation, even exceeding the network structure's influence in some cases. These observations present a challenge to the conventional wisdom of network architectures in this stage. We expect these discoveries to encourage researchers to rethink the current paradigm of "excessive attention on the performance of a single small dataset" in stereo matching.

Determination of acyclovir in renal microdialysis fluid and confirmation of renal function index.

Acyclovir (ACV) is a nucleoside antivirus-free agent that was developed and marketed by Burroughs Well-come of the United States. Renal damage from ACV has been a major factor limiting its clinical application. Thus, the renal toxicity mechanism of ACV requires systematic study. In our previous study, we speculated that the nephrotoxicity of ACV may be associated with oxidative stress. In addition to the study of ACV's toxic effect in vivo, it is also necessary to explore the absorption and distribution of ACV in the body to further investigate the changes to ACV in the body. In this study, the toxicokinetics ACV in the kidney of the rat were explored using microdialysis, and the renal function of rats was measured. The results showed that high-dose ACV is associated with renal toxicity after a single intravenous injection or successive administration.

Osthole induced apoptosis in human normal liver cells by regulating cell proliferation and endoplasmic reticulum stress.

Osthole (Ost) is often used in treatment for cancer, inflammation and rheumatism in clinic. However, Ost-induced liver injury has been reported. In this study, we aim to investigate the possible mechanism of Ost-induced hepatotoxicity in human normal liver cells (L02). When cells were exposed to Ost, the cell viability was decreased and apoptosis rate increased, the intracellular markers of oxidative stress were changed. Simultaneously, Ost altered apoptotic related proteins levels, including Bcl-2, Bax, Cleaved-Caspase-9/-8/-3, and Pro-Caspase-3/-8. In addition, Ost enhanced the levels of endoplasmic reticulum (ER) stress proteins (GRP78/Bip, CHOP, Caspase-4, IRE1α, PERK, JNK, P-JNK, and ATF4), decreased the cell proliferation and cycle-associated protein (Phospho-Histone H3, P-Cdc25C, Cdc25C, P-Cdc2, Cdc2, and Cyclin B1) level. The results show that Ost has toxic effects on L02 cells. Furthermore, it induces apoptosis by inhibiting cell proliferation, arresting cell cycle at the G2/M phase and activating ER stress.

Oxymatrine Causes Hepatotoxicity by Promoting the Phosphorylation of JNK and Induction of Endoplasmic Reticulum Stress Mediated by ROS in LO2 Cells.

Oxymatrine (OMT) often used in treatment for chronic hepatitis B virus infection in clinic. However, OMT-induced liver injury has been reported. In this study, we aim to investigate the possible mechanism of OMT-induced hepatotoxicity in human normal liver cells (L02). Exposed cells to OMT, the cell viability was decreased and apoptosis rate increased, the intracellular markers of oxidative stress were changed. Simultaneously, OMT altered apoptotic related proteins levels, including Bcl-2, Bax and pro-caspase-8/-9/-3. In addition, OMT enhanced the protein levels of endoplasmic reticulum (ER) stress makers (GRP78/Bip, CHOP, and cleaved-Caspase-4) and phosphorylation of c-Jun N-terminal kinase (p-JNK), as well as the mRNA levels of GRP78/Bip, CHOP, caspase-4, and ER stress sensors (IREI, ATF6, and PERK). Pre-treatment with Z-VAD-fmk, JNK inhibitor SP600125 and N-acetyl-l-cysteine (NAC), a ROS scavenger, partly improved the survival rates and restored OMT-induced cellular damage, and reduced caspase-3 cleavage. SP600125 or NAC reduced OMT-induced p-JNK and NAC significantly lowered caspase-4. Furthermore, 4-PBA, the ER stress inhibitor, weakened inhibitory effect of OMT on cells, on the contrary, TM worsen. 4-PBA also reduced the levels of p-JNK and cleaved-caspase-3 proteins. Therefore, OMT-induced injury in L02 cells was related to ROS mediated p-JNK and ER stress induction. Antioxidant, by inhibition of p-JNK or ER stress, may be a feasible method to alleviate OMT-induced liver injury.