NSTL国家科技图书文献中心

1. SCOMatch: Alleviating Overtrusting in Open-Set Semi-supervised Learning NSTL国家科技图书文献中心

Zerun Wang | Liuyu Xiang ... - 《Computer Vision - ECCV 2024,Part LI》 - European Conference on Computer Vision - 2025, - 217～233 - 共17页

摘要：Open-set semi-supervised* learning (OSSL* | classes, for semi-supervised* learning (SSL). Prior OSSL* | ) leverages practical open-set unlabeled data, comprising | both in-distribution (ID) samples from seen classes | and out-of-distribution (OOD) samples from unseen

关键词： Open-set problem | Semi-supervised learning

2. Full-Body Human De-lighting with Semi-supervised Learning NSTL国家科技图书文献中心

Joshua Weir | Junhong Zhao ... - 《Computer Vision - ACCV 2024,Part I》 - Asian Conference on Computer Vision - 2025, - 165～181 - 共17页

摘要： novel semi-supervised* deep learning method to* | applications. While recent advancements in deep learning | Removing undesired shading from human images | is crucial in supporting various real-world | -based methods show promise in addressing this

关键词： De-lighting | Full-body | Semi-supervised learning

3. ProSub: Probabilistic Open-Set Semi-supervised Learning with Subspace-Based Out-of-Distribution Detection NSTL国家科技图书文献中心

Erik Wallin | Lennart Svensson ... - 《Computer Vision - ECCV 2024,Part LXI》 - European Conference on Computer Vision - 2025, - 129～147 - 共19页

摘要：In open-set semi-supervised* learning (OSSL* | ), we consider unlabeled datasets that may contain | unknown classes. Existing OSSL methods often use the | softmax confidence for classifying data as in | -distribution (ID) or out-of-distribution (OOD). Additionally

关键词： Open-set semi-supervised learning

4. Improving 3D Semi-supervised Learning by Effectively Utilizing All Unlabelled Data NSTL国家科技图书文献中心

Sneha Paul | Zachary Patterson ... - 《Computer Vision - ECCV 2024,Part LXVI》 - European Conference on Computer Vision - 2025, - 55～71 - 共17页

摘要：Semi-supervised learning (SSL) has shown its | unlabelled data. Traditional semi-supervised* approaches* | effectiveness in learning* effective 3D representation from a* | the learning* process. However, we identify that the* | such samples, (2) an inverse learning* module that*

关键词： Semi-supervised learning | Point cloud | 3D vision

5. Flexible Distribution Alignment: Towards Long-Tailed Semi-supervised Learning with Proper Calibration NSTL国家科技图书文献中心

Emanuel Sanchez Aima... | Nathaniel Helgesen ... - 《Computer Vision - ECCV 2024,Part LIV》 - European Conference on Computer Vision - 2025, - 307～327 - 共21页

摘要：Long-tailed semi-supervised* learning (LTSSL* | ) represents a practical scenario for semi-supervised | -supervised* learning. Our code is available at https* | ImageNet127, addressing class imbalance challenges in semi | applications, challenged by skewed labeled distributions that

关键词： Distribution alignment | Confidence calibration | Long-tailed | Semi-supervised learning

6. All-Weather Vehicle Detection and Classification with Adversarial and Semi-Supervised Learning NSTL国家科技图书文献中心

Yi-Chao Huang | Huei-Yung Lin - 《Pattern Recognition,Part XVII》 - International Conference on Pattern Recognition - 2025, - 330～345 - 共16页

摘要： semi-supervised learning framework is incorporated to | learning to ensure the features extracted from a similar | learning technique outperforms the recent traffic scene | The images taken under varying lighting or | adverse weather conditions exhibit different

关键词： All weather object detection | Adversarial learning | Semi-Supervised learning

7. ExMatch: Self-guided Exploitation for Semi-supervised Learning with Scarce Labeled Samples NSTL国家科技图书文献中心

Noo-ri Kim | Jin-Seop Lee ... - 《Computer Vision - ECCV 2024,Part LXXXV》 - European Conference on Computer Vision - 2025, - 125～142 - 共18页

摘要：Semi-supervised learning is a learning method | labeled samples, semi-supervised* learning methods showed* | semi-supervised learning. In the training process | samples using self-supervised* models and utilize it for* | distribution and resist learning* from incorrect pseudo-labels*

关键词： Semi-supervised learning | Scarce labeled samples | Self-guided exploitation | Confirmation bias

8. Rebalancing Using Estimated Class Distribution for Imbalanced Semi-supervised Learning Under Class Distribution Mismatch NSTL国家科技图书文献中心

Taemin Park | Hyuck Lee ... - 《Computer Vision - ECCV 2024,Part XXII》 - European Conference on Computer Vision - 2025, - 388～404 - 共17页

摘要：-imbalanced semi-supervised* learning (CISSL), many existing* | Despite significant advancements in class | algorithms explicitly or implicitly assume that the class | distribution of unlabeled data matches that of labeled data | . However, when this assumption fails in practice, the

关键词： Class-imbalanced semi-supervised learning | Long-tailed learning | Auxiliary balanced classifier

9. Essay Coherence Evaluation and Feedback Enhanced by Semi-supervised Learning and Auxiliary Information NSTL国家科技图书文献中心

Chenyang Li | Long Zhang ... - 《Natural Language Processing and Chinese Computing,Part V》 - CCF International Conference on Natural Language Processing and Chinese Computing - 2025, - 312～321 - 共10页

摘要： generated through semi-supervised* learning and information* | According to the grading standards for middle | school examinations, thought content and structural | coherence are critical indicators of essay quality | . Evaluating an essay's coherence requires not only analyzing

关键词： Essay quality assessment | Semi-Supervised learning | Large language models | Auxiliary data

10. Self-training Based Semi-Supervised Learning and U-Net with Denoiser for Teeth Segmentation in X-Ray Image NSTL国家科技图书文献中心

Zhouhao Lin | Yibo Yang ... - 《Semi-supervised Tooth Segmentation》 - MICCAI Challenge on Semi-supervised Tooth Segmentation | International Conference on Medical Image Computing and Computer Assisted Intervention - 2025, - 124～132 - 共9页

摘要：Segmentation of X-ray images of teeth plays an | important role in dental diagnosis. However, the labeling | information is expensive, and it is of great significance to | effectively use the unlabeled images to improve the | segmentation performance. In order to make full use of

摘要：Open-set semi-supervised* learning (OSSL* | classes, for semi-supervised* learning (SSL). Prior OSSL* | ) leverages practical open-set unlabeled data, comprising | both in-distribution (ID) samples from seen classes | and out-of-distribution (OOD) samples from unseen

关键词： Open-set problem | Semi-supervised learning

摘要： novel semi-supervised* deep learning method to* | applications. While recent advancements in deep learning | Removing undesired shading from human images | is crucial in supporting various real-world | -based methods show promise in addressing this

关键词： De-lighting | Full-body | Semi-supervised learning

摘要：In open-set semi-supervised* learning (OSSL* | ), we consider unlabeled datasets that may contain | unknown classes. Existing OSSL methods often use the | softmax confidence for classifying data as in | -distribution (ID) or out-of-distribution (OOD). Additionally

关键词： Open-set semi-supervised learning

摘要：Semi-supervised learning (SSL) has shown its | unlabelled data. Traditional semi-supervised* approaches* | effectiveness in learning* effective 3D representation from a* | the learning* process. However, we identify that the* | such samples, (2) an inverse learning* module that*

关键词： Semi-supervised learning | Point cloud | 3D vision

摘要：Long-tailed semi-supervised* learning (LTSSL* | ) represents a practical scenario for semi-supervised | -supervised* learning. Our code is available at https* | ImageNet127, addressing class imbalance challenges in semi | applications, challenged by skewed labeled distributions that

关键词： Distribution alignment | Confidence calibration | Long-tailed | Semi-supervised learning

摘要： semi-supervised learning framework is incorporated to | learning to ensure the features extracted from a similar | learning technique outperforms the recent traffic scene | The images taken under varying lighting or | adverse weather conditions exhibit different

关键词： All weather object detection | Adversarial learning | Semi-Supervised learning

摘要：Semi-supervised learning is a learning method | labeled samples, semi-supervised* learning methods showed* | semi-supervised learning. In the training process | samples using self-supervised* models and utilize it for* | distribution and resist learning* from incorrect pseudo-labels*

关键词： Semi-supervised learning | Scarce labeled samples | Self-guided exploitation | Confirmation bias

摘要：-imbalanced semi-supervised* learning (CISSL), many existing* | Despite significant advancements in class | algorithms explicitly or implicitly assume that the class | distribution of unlabeled data matches that of labeled data | . However, when this assumption fails in practice, the

关键词： Class-imbalanced semi-supervised learning | Long-tailed learning | Auxiliary balanced classifier

摘要： generated through semi-supervised* learning and information* | According to the grading standards for middle | school examinations, thought content and structural | coherence are critical indicators of essay quality | . Evaluating an essay's coherence requires not only analyzing

关键词： Essay quality assessment | Semi-Supervised learning | Large language models | Auxiliary data

摘要：Segmentation of X-ray images of teeth plays an | important role in dental diagnosis. However, the labeling | information is expensive, and it is of great significance to | effectively use the unlabeled images to improve the | segmentation performance. In order to make full use of

关键词： Semi-supervised learning | Teeth segmentation | Denoiser | Hard augmentation

4008-161-200 800-990-8900

国家科技图书文献中心

摘要：Open-set semi-supervised learning (OSSL | classes, for semi-supervised learning (SSL). Prior OSSL | ) leverages practical open-set unlabeled data, comprising | both in-distribution (ID) samples from seen classes | and out-of-distribution (OOD) samples from unseen

关键词： Open-set problem | Semi-supervised learning

摘要： novel semi-supervised deep learning method to | applications. While recent advancements in deep learning | Removing undesired shading from human images | is crucial in supporting various real-world | -based methods show promise in addressing this

关键词： De-lighting | Full-body | Semi-supervised learning

摘要：In open-set semi-supervised learning (OSSL | ), we consider unlabeled datasets that may contain | unknown classes. Existing OSSL methods often use the | softmax confidence for classifying data as in | -distribution (ID) or out-of-distribution (OOD). Additionally

关键词： Open-set semi-supervised learning

摘要：Semi-supervised learning (SSL) has shown its | unlabelled data. Traditional semi-supervised approaches | effectiveness in learning effective 3D representation from a | the learning process. However, we identify that the | such samples, (2) an inverse learning module that

关键词： Semi-supervised learning | Point cloud | 3D vision

摘要：Long-tailed semi-supervised learning (LTSSL | ) represents a practical scenario for semi-supervised | -supervised learning. Our code is available at https | ImageNet127, addressing class imbalance challenges in semi | applications, challenged by skewed labeled distributions that

关键词： Distribution alignment | Confidence calibration | Long-tailed | Semi-supervised learning

摘要： semi-supervised learning framework is incorporated to | learning to ensure the features extracted from a similar | learning technique outperforms the recent traffic scene | The images taken under varying lighting or | adverse weather conditions exhibit different

关键词： All weather object detection | Adversarial learning | Semi-Supervised learning

摘要：Semi-supervised learning is a learning method | labeled samples, semi-supervised learning methods showed | semi-supervised learning. In the training process | samples using self-supervised models and utilize it for | distribution and resist learning from incorrect pseudo-labels

关键词： Semi-supervised learning | Scarce labeled samples | Self-guided exploitation | Confirmation bias

摘要：-imbalanced semi-supervised learning (CISSL), many existing | Despite significant advancements in class | algorithms explicitly or implicitly assume that the class | distribution of unlabeled data matches that of labeled data | . However, when this assumption fails in practice, the

关键词： Class-imbalanced semi-supervised learning | Long-tailed learning | Auxiliary balanced classifier

摘要： generated through semi-supervised learning and information | According to the grading standards for middle | school examinations, thought content and structural | coherence are critical indicators of essay quality | . Evaluating an essay's coherence requires not only analyzing

关键词： Essay quality assessment | Semi-Supervised learning | Large language models | Auxiliary data

摘要：Segmentation of X-ray images of teeth plays an | important role in dental diagnosis. However, the labeling | information is expensive, and it is of great significance to | effectively use the unlabeled images to improve the | segmentation performance. In order to make full use of

关键词： Semi-supervised learning | Teeth segmentation | Denoiser | Hard augmentation

4008-161-200 800-990-8900

国家科技图书文献中心

摘要：Open-set semi-supervised* learning (OSSL* | classes, for semi-supervised* learning (SSL). Prior OSSL* | ) leverages practical open-set unlabeled data, comprising | both in-distribution (ID) samples from seen classes | and out-of-distribution (OOD) samples from unseen

摘要： novel semi-supervised* deep learning method to* | applications. While recent advancements in deep learning | Removing undesired shading from human images | is crucial in supporting various real-world | -based methods show promise in addressing this

摘要：In open-set semi-supervised* learning (OSSL* | ), we consider unlabeled datasets that may contain | unknown classes. Existing OSSL methods often use the | softmax confidence for classifying data as in | -distribution (ID) or out-of-distribution (OOD). Additionally

摘要：Semi-supervised learning (SSL) has shown its | unlabelled data. Traditional semi-supervised* approaches* | effectiveness in learning* effective 3D representation from a* | the learning* process. However, we identify that the* | such samples, (2) an inverse learning* module that*

摘要：Long-tailed semi-supervised* learning (LTSSL* | ) represents a practical scenario for semi-supervised | -supervised* learning. Our code is available at https* | ImageNet127, addressing class imbalance challenges in semi | applications, challenged by skewed labeled distributions that

摘要：Semi-supervised learning is a learning method | labeled samples, semi-supervised* learning methods showed* | semi-supervised learning. In the training process | samples using self-supervised* models and utilize it for* | distribution and resist learning* from incorrect pseudo-labels*

摘要：-imbalanced semi-supervised* learning (CISSL), many existing* | Despite significant advancements in class | algorithms explicitly or implicitly assume that the class | distribution of unlabeled data matches that of labeled data | . However, when this assumption fails in practice, the

摘要： generated through semi-supervised* learning and information* | According to the grading standards for middle | school examinations, thought content and structural | coherence are critical indicators of essay quality | . Evaluating an essay's coherence requires not only analyzing