The Experts below are selected from a list of 32949 Experts worldwide ranked by ideXlab platform
Philip H S Torr - One of the best experts on this subject based on the ideXlab platform.
-
domain invariant stereo matching networks
European Conference on Computer Vision, 2020Co-Authors: Feihu Zhang, Ruigang Yang, Victor Adrian Prisacariu, Benjamin W Wah, Philip H S TorrAbstract:State-of-the-art stereo matching networks have difficulties in generalizing to new unseen environments due to significant domain differences, such as color, illumination, contrast, and texture. In this paper, we aim at designing a domain-invariant stereo matching network (DSMNet) that generalizes well to unseen scenes. To achieve this goal, we propose i) a novel “domain Normalization” Approach that regularizes the distribution of learned representations to allow them to be invariant to domain differences, and ii) an end-to-end trainable structure-preserving graph-based filter for extracting robust structural and geometric representations that can further enhance domain-invariant generalizations. When trained on synthetic data and generalized to real test sets, our model performs significantly better than all state-of-the-art models. It even outperforms some deep neural network models (e.g. MC-CNN [61]) fine-tuned with test-domain data. The code is available at https://github.com/feihuzhang/DSMNet.
-
domain invariant stereo matching networks
arXiv: Computer Vision and Pattern Recognition, 2019Co-Authors: Feihu Zhang, Ruigang Yang, Victor Adrian Prisacariu, Benjamin W Wah, Philip H S TorrAbstract:State-of-the-art stereo matching networks have difficulties in generalizing to new unseen environments due to significant domain differences, such as color, illumination, contrast, and texture. In this paper, we aim at designing a domain-invariant stereo matching network (DSMNet) that generalizes well to unseen scenes. To achieve this goal, we propose i) a novel "domain Normalization" Approach that regularizes the distribution of learned representations to allow them to be invariant to domain differences, and ii) a trainable non-local graph-based filter for extracting robust structural and geometric representations that can further enhance domain-invariant generalizations. When trained on synthetic data and generalized to real test sets, our model performs significantly better than all state-of-the-art models. It even outperforms some deep learning models (e.g. MC-CNN) fine-tuned with test-domain data.
Wei Wang - One of the best experts on this subject based on the ideXlab platform.
-
automatic analysis system of calcaneus radiograph rotation invariant landmark detection for calcaneal angle measurement fracture identification and fracture region segmentation
Computer Methods and Programs in Biomedicine, 2021Co-Authors: Jia Guo, Dong Xue, Junxian Chen, Huanxin Yan, Wei WangAbstract:Abstract Background and objective Calcaneus is the largest tarsal bone to withstand the daily stresses of weight-bearing. The calcaneal fracture is the most common type in the tarsal bone fractures. After a fracture is suspected, plain radiographs should be taken first. Bohler's Angle (BA) and Critical Angle of Gissane (CAG), measured by four anatomic landmarks in lateral foot radiograph, can guide fracture diagnosis and facilitate operative recovery of the fractured calcaneus. This study aims to develop an analysis system that can automatically locate four anatomic landmarks, measure BA and CAG for fracture assessment, identify fractured calcaneus, and segment fractured regions. Methods For landmark detection, we proposed a coarse-to-fine Rotation-Invariant Regression-Voting (RIRV) landmark detection method based on regressive Multi-Layer Perceptron (MLP) and Scale Invariant Feature Transform (SIFT) patch descriptor, which solves the problem of fickle rotation of calcaneus. By implementing a novel Normalization Approach, the RIRV method is explicitly rotation-invariance comparing with traditional regressive methods. For fracture identification and segmentation, a convolution neural network (CNN) based on U-Net with auxiliary classification head (U-Net-CH) is designed. The input ROIs of the CNN are normalized by detected landmarks to uniform view, orientation, and scale. The advantage of this Approach is the multi-task learning that combines classification and segmentation. Results Our system can accurately measure BA and CAG with a mean angle error of 3.8○ and 6.2○ respectively. For fracture identification and fracture region segmentation, our system presents good performance with an F1-score of 96.55%, recall of 94.99%, and segmentation IoU-score of 0.586. Conclusion A powerful calcaneal radiograph analysis system including anatomical angles measurement, fracture identification, and fracture segmentation can be built. The proposed analysis system can aid orthopedists to improve the efficiency and accuracy of calcaneus fracture diagnosis.
Graham Merrington - One of the best experts on this subject based on the ideXlab platform.
-
does the scientific underpinning of regulatory tools to estimate bioavailability of nickel in freshwaters matter the european wide environmental quality standard for nickel
Environmental Toxicology and Chemistry, 2016Co-Authors: Adam Peters, Christian E Schlekat, Graham MerringtonAbstract:: A bioavailability-based environmental quality standard (EQS) was established for nickel in freshwaters under the European Union's Water Framework Directive. Bioavailability correction based on pH, water hardness, and dissolved organic carbon is a demonstrable improvement on existing hardness-based quality standards, which may be underprotective in high-hardness waters. The present study compares several simplified bioavailability tools developed to implement the Ni EQS (biomet, M-BAT, and PNECPro) against the full bioavailability Normalization procedure on which the EQS was based. Generally, all tools correctly distinguished sensitive waters from insensitive waters, although with varying degrees of accuracy compared with full Normalization. Biomet and M-BAT predictions were consistent with, but less accurate than, full bioavailability Normalization results, whereas PNECpro results were generally more conservative. The comparisons revealed important differences in tools in development, which results in differences in the predictions. Importantly, the models used for the development of PNECpro use a different ecotoxicity dataset, and a different bioavailability Normalization Approach using fewer biotic ligand models (BLMs) than that used for the derivation of the Ni EQS. The failure to include all of the available toxicity data, and all of the appropriate NiBLMs, has led to some significant differences between the predictions provided by PNECpro and those calculated using the process agreed to in Europe under the Water Framework Directive and other chemicals management programs (such as REACH). These considerable differences mean that PNECpro does not reflect the behavior, fate, and ecotoxicity of nickel, and raises concerns about its applicability for checking compliance against the Ni EQS. Environ Toxicol Chem 2016;35:2397-2404. © 2016 SETAC.
-
does the scientific underpinning of regulatory tools to estimate bioavailability of nickel in freshwaters matter the european wide environmental quality standard for nickel
Environmental Toxicology and Chemistry, 2016Co-Authors: Adam Peters, Christian E Schlekat, Graham MerringtonAbstract:: A bioavailability-based environmental quality standard (EQS) was established for nickel in freshwaters under the European Union's Water Framework Directive. Bioavailability correction based on pH, water hardness, and dissolved organic carbon is a demonstrable improvement on existing hardness-based quality standards, which may be underprotective in high-hardness waters. The present study compares several simplified bioavailability tools developed to implement the Ni EQS (biomet, M-BAT, and PNECPro) against the full bioavailability Normalization procedure on which the EQS was based. Generally, all tools correctly distinguished sensitive waters from insensitive waters, although with varying degrees of accuracy compared with full Normalization. Biomet and M-BAT predictions were consistent with, but less accurate than, full bioavailability Normalization results, whereas PNECpro results were generally more conservative. The comparisons revealed important differences in tools in development, which results in differences in the predictions. Importantly, the models used for the development of PNECpro use a different ecotoxicity dataset, and a different bioavailability Normalization Approach using fewer biotic ligand models (BLMs) than that used for the derivation of the Ni EQS. The failure to include all of the available toxicity data, and all of the appropriate NiBLMs, has led to some significant differences between the predictions provided by PNECpro and those calculated using the process agreed to in Europe under the Water Framework Directive and other chemicals management programs (such as REACH). These considerable differences mean that PNECpro does not reflect the behavior, fate, and ecotoxicity of nickel, and raises concerns about its applicability for checking compliance against the Ni EQS. Environ Toxicol Chem 2016;35:2397-2404. © 2016 SETAC.
Feihu Zhang - One of the best experts on this subject based on the ideXlab platform.
-
domain invariant stereo matching networks
European Conference on Computer Vision, 2020Co-Authors: Feihu Zhang, Ruigang Yang, Victor Adrian Prisacariu, Benjamin W Wah, Philip H S TorrAbstract:State-of-the-art stereo matching networks have difficulties in generalizing to new unseen environments due to significant domain differences, such as color, illumination, contrast, and texture. In this paper, we aim at designing a domain-invariant stereo matching network (DSMNet) that generalizes well to unseen scenes. To achieve this goal, we propose i) a novel “domain Normalization” Approach that regularizes the distribution of learned representations to allow them to be invariant to domain differences, and ii) an end-to-end trainable structure-preserving graph-based filter for extracting robust structural and geometric representations that can further enhance domain-invariant generalizations. When trained on synthetic data and generalized to real test sets, our model performs significantly better than all state-of-the-art models. It even outperforms some deep neural network models (e.g. MC-CNN [61]) fine-tuned with test-domain data. The code is available at https://github.com/feihuzhang/DSMNet.
-
domain invariant stereo matching networks
arXiv: Computer Vision and Pattern Recognition, 2019Co-Authors: Feihu Zhang, Ruigang Yang, Victor Adrian Prisacariu, Benjamin W Wah, Philip H S TorrAbstract:State-of-the-art stereo matching networks have difficulties in generalizing to new unseen environments due to significant domain differences, such as color, illumination, contrast, and texture. In this paper, we aim at designing a domain-invariant stereo matching network (DSMNet) that generalizes well to unseen scenes. To achieve this goal, we propose i) a novel "domain Normalization" Approach that regularizes the distribution of learned representations to allow them to be invariant to domain differences, and ii) a trainable non-local graph-based filter for extracting robust structural and geometric representations that can further enhance domain-invariant generalizations. When trained on synthetic data and generalized to real test sets, our model performs significantly better than all state-of-the-art models. It even outperforms some deep learning models (e.g. MC-CNN) fine-tuned with test-domain data.
Jia Guo - One of the best experts on this subject based on the ideXlab platform.
-
automatic analysis system of calcaneus radiograph rotation invariant landmark detection for calcaneal angle measurement fracture identification and fracture region segmentation
Computer Methods and Programs in Biomedicine, 2021Co-Authors: Jia Guo, Dong Xue, Junxian Chen, Huanxin Yan, Wei WangAbstract:Abstract Background and objective Calcaneus is the largest tarsal bone to withstand the daily stresses of weight-bearing. The calcaneal fracture is the most common type in the tarsal bone fractures. After a fracture is suspected, plain radiographs should be taken first. Bohler's Angle (BA) and Critical Angle of Gissane (CAG), measured by four anatomic landmarks in lateral foot radiograph, can guide fracture diagnosis and facilitate operative recovery of the fractured calcaneus. This study aims to develop an analysis system that can automatically locate four anatomic landmarks, measure BA and CAG for fracture assessment, identify fractured calcaneus, and segment fractured regions. Methods For landmark detection, we proposed a coarse-to-fine Rotation-Invariant Regression-Voting (RIRV) landmark detection method based on regressive Multi-Layer Perceptron (MLP) and Scale Invariant Feature Transform (SIFT) patch descriptor, which solves the problem of fickle rotation of calcaneus. By implementing a novel Normalization Approach, the RIRV method is explicitly rotation-invariance comparing with traditional regressive methods. For fracture identification and segmentation, a convolution neural network (CNN) based on U-Net with auxiliary classification head (U-Net-CH) is designed. The input ROIs of the CNN are normalized by detected landmarks to uniform view, orientation, and scale. The advantage of this Approach is the multi-task learning that combines classification and segmentation. Results Our system can accurately measure BA and CAG with a mean angle error of 3.8○ and 6.2○ respectively. For fracture identification and fracture region segmentation, our system presents good performance with an F1-score of 96.55%, recall of 94.99%, and segmentation IoU-score of 0.586. Conclusion A powerful calcaneal radiograph analysis system including anatomical angles measurement, fracture identification, and fracture segmentation can be built. The proposed analysis system can aid orthopedists to improve the efficiency and accuracy of calcaneus fracture diagnosis.