NSTL国家科技图书文献中心

1. Efficient Lung Segmentation for Tumour Detection NSTL国家科技图书文献中心

Anura Hiraman | Serestina Viriri ... - 《Advances in Visual Computing,Part I》 - International Symposium on Visual Computing - 2025, - 507～518 - 共12页

摘要：Over the past decade, deep learning has | significantly impacted medical imaging, particularly in | segmentation tasks. Encoder-decoder methods have advanced | medical image segmentation by restoring feature map | resolution and minimizing information loss during decoding

关键词： CNN | Encoder-Decoder | Lung segmentation | Region of interest

2. AFIDAF: Alternating Fourier and Image Domain Adaptive Filters as an Efficient Alternative to Attention in ViTs NSTL国家科技图书文献中心

Yunling Zheng | Zeyi Xu ... - 《Advances in Visual Computing,Part I》 - International Symposium on Visual Computing - 2025, - 17～30 - 共14页

摘要：We propose and demonstrate an alternating | Fourier and image domain filtering approach for feature | extraction as an efficient alternative to build a vision | backbone without using the computationally intensive | attention. The performance among the lightweight models

关键词： Fourier domain filtering | Group shuffled large kernel convolution | Dual domain feature extraction

3. Human Pose Estimation-Based ID Assignment Method in the Wild: A Real-Time Approach NSTL国家科技图书文献中心

Pyunghwa Shin | Ohung Kwon - 《Advances in Visual Computing,Part I》 - International Symposium on Visual Computing - 2025, - 223～236 - 共14页

摘要：In the field of computer vision, research on | human pose estimation aims to understand human | movements by analyzing individual camera images. However | , identifying individuals in multi-people pose estimation | remains a challenging issue. This study explores ID

关键词： ID assignment | Human pose estimation | Real-Time image analysis | In the wild environments

4. ActiveConfusion: Efficient Cold-Start Active Learning Using Pretext Task Confusion NSTL国家科技图书文献中心

Azeez Idris | Wallapak Tavanapong - 《Advances in Visual Computing,Part I》 - International Symposium on Visual Computing - 2025, - 357～371 - 共15页

摘要：Active Learning (AL) reduces manual annotation | costs by iteratively selecting a small, but most | informative subset of an unlabeled dataset for labeling by | domain experts. The quality of the initial subset can | significantly impact the entire AL process. The problem of

关键词： Active learning | Cold-Start problem | Self-Supervised learning | Confusion score

5. Estimation of Global Illumination Using Cycle-Consistent Adversarial Networks NSTL国家科技图书文献中心

Junho Oh | Amos Lynn Abbott - 《Advances in Visual Computing,Part I》 - International Symposium on Visual Computing - 2025, - 73～86 - 共14页

摘要：Synthesis of realistic virtual environments | requires careful rendering of light and shadows, a task | often bottle-necked by the high computational cost of | global illumination (GI) techniques. This paper | introduces a new GI approach that improves computational

关键词： Virtual reality | Video games | Graphics | Lighting | Global illumination | GAN | CycleGAN

6. Hybrid Voxel Formats for Efficient Ray Tracing NSTL国家科技图书文献中心

Russel Arbore | Jeffrey Liu ... - 《Advances in Visual Computing,Part I》 - International Symposium on Visual Computing - 2025, - 125～138 - 共14页

摘要：Voxels are a geometric representation used for | rendering volumes, multi-resolution models, and indirect | lighting effects. Since the memory consumption of | uncompressed voxel volumes scales cubically with resolution | , past works have introduced data structures for

关键词： Voxels | Ray tracing | Compression | Metaprogramming

7. Visualizing Polarization Effects of Gravitational Waves Using Particle Rings and Surfaces in Virtual Reality NSTL国家科技图书文献中心

Kristen Schumacher | Sonali Joshi ... - 《Advances in Visual Computing,Part I》 - International Symposium on Visual Computing - 2025, - 413～424 - 共12页

摘要：In some modified theories of gravity | , gravitational waves may contain up to six polarizations. These | different polarizations determine the pattern with which | the gravitational waves stretch and squeeze | spacetime, both perpendicular to and along the direction

关键词： Visualization | Physical sciences | Gravitational waves | VR

8. PIEPredict++: An Improved Pedestrian Intention Estimation Model Incorporating Comprehensive Environment Information NSTL国家科技图书文献中心

Yikai Bao | Yukihiro Saito ... - 《Advances in Visual Computing,Part I》 - International Symposium on Visual Computing - 2025, - 141～155 - 共15页

摘要：Predicting pedestrians' highly variable | crossing intentions has attracted considerable attention | in autonomous driving research. This paper | introduces PIEPredict++, a new model for predicting | pedestrian intentions. It uses dynamic and static

关键词： Pedestrian intention estimation | Autonomous driving | Perception | Deep learning | Cross-attention

9. Seeing Is Believing: The Role of Scatterplots in Recommender System Trust and Decision-Making NSTL国家科技图书文献中心

Bhavana Doppalapudi | Md Dilshadur Rahman ... - 《Advances in Visual Computing,Part I》 - International Symposium on Visual Computing - 2025, - 425～438 - 共14页

摘要：The accuracy of recommender systems influences | their trust and decision-making when using them | . Providing additional information, such as visualizations | , offers context that would otherwise be lacking. However | , the role of visualizations in influencing trust and

关键词： Machine learning | Visualizations | Trust | Decision-Making

10. Explainable Action-Recognition Based Approach for Unsupervised Video Anomaly Detection NSTL国家科技图书文献中心

Hui Yang | Mostepha Khouadjia ... - 《Advances in Visual Computing,Part I》 - International Symposium on Visual Computing - 2025, - 182～195 - 共14页

摘要：In recent years, there has been a growing | focus on video anomaly detection from unlabeled data | , raising the task of Unsupervised Video Abnormal | Detection (UVAD). However, most of the existing approaches | predominantly focus on global features derived from entire

摘要：Over the past decade, deep learning has | significantly impacted medical imaging, particularly in | segmentation tasks. Encoder-decoder methods have advanced | medical image segmentation by restoring feature map | resolution and minimizing information loss during decoding

关键词： CNN | Encoder-Decoder | Lung segmentation | Region of interest

摘要：We propose and demonstrate an alternating | Fourier and image domain filtering approach for feature | extraction as an efficient alternative to build a vision | backbone without using the computationally intensive | attention. The performance among the lightweight models

关键词： Fourier domain filtering | Group shuffled large kernel convolution | Dual domain feature extraction

摘要：In the field of computer vision, research on | human pose estimation aims to understand human | movements by analyzing individual camera images. However | , identifying individuals in multi-people pose estimation | remains a challenging issue. This study explores ID

关键词： ID assignment | Human pose estimation | Real-Time image analysis | In the wild environments

摘要：Active Learning (AL) reduces manual annotation | costs by iteratively selecting a small, but most | informative subset of an unlabeled dataset for labeling by | domain experts. The quality of the initial subset can | significantly impact the entire AL process. The problem of

关键词： Active learning | Cold-Start problem | Self-Supervised learning | Confusion score

摘要：Synthesis of realistic virtual environments | requires careful rendering of light and shadows, a task | often bottle-necked by the high computational cost of | global illumination (GI) techniques. This paper | introduces a new GI approach that improves computational

关键词： Virtual reality | Video games | Graphics | Lighting | Global illumination | GAN | CycleGAN

摘要：Voxels are a geometric representation used for | rendering volumes, multi-resolution models, and indirect | lighting effects. Since the memory consumption of | uncompressed voxel volumes scales cubically with resolution | , past works have introduced data structures for

关键词： Voxels | Ray tracing | Compression | Metaprogramming

摘要：In some modified theories of gravity | , gravitational waves may contain up to six polarizations. These | different polarizations determine the pattern with which | the gravitational waves stretch and squeeze | spacetime, both perpendicular to and along the direction

关键词： Visualization | Physical sciences | Gravitational waves | VR

摘要：Predicting pedestrians' highly variable | crossing intentions has attracted considerable attention | in autonomous driving research. This paper | introduces PIEPredict++, a new model for predicting | pedestrian intentions. It uses dynamic and static

关键词： Pedestrian intention estimation | Autonomous driving | Perception | Deep learning | Cross-attention

摘要：The accuracy of recommender systems influences | their trust and decision-making when using them | . Providing additional information, such as visualizations | , offers context that would otherwise be lacking. However | , the role of visualizations in influencing trust and

关键词： Machine learning | Visualizations | Trust | Decision-Making

摘要：In recent years, there has been a growing | focus on video anomaly detection from unlabeled data | , raising the task of Unsupervised Video Abnormal | Detection (UVAD). However, most of the existing approaches | predominantly focus on global features derived from entire

关键词： Video anomaly detection | Explainability | Unsupervised learning

4008-161-200 800-990-8900

国家科技图书文献中心