全部 |
  • 全部
  • 题名
  • 作者
  • 机构
  • 关键词
  • NSTL主题词
  • 摘要
检索 二次检索 AI检索
外文文献 中文文献
筛选条件:

1. Vector Quantized Underwater Image Enhancement With Transformers NSTL国家科技图书文献中心

Xueyan Ding |  Yixin Sui... -  《IEEE Journal of Oceanic Engineering》 - 2025,50(1) - 136~149 - 共14页

摘要: representations of underwater images through a vector quantized | , we propose a vector quantized underwater image |  vector quantized generative network and an axial flow | -guided latent transformer. The vector quantized | Due to the complexity of underwater imaging
关键词: Vectors |  Transformers |  Image color analysis |  Codes |  Feature extraction |  Image enhancement |  Convolutional neural networks |  Complexity theory |  Image reconstruction |  Generative adversarial networks

2. Quaternion Vector Quantized Variational Autoencoder NSTL国家科技图书文献中心

Hui Luo |  Xin Liu... -  《IEEE signal processing letters》 - 2025,32 - 151~155 - 共5页

摘要:Vector quantized variational autoencoders, as |  discrete representations by quantizing continuous latent |  variants of variational autoencoders, effectively capture |  spaces and are widely used in generative tasks. However | , these models still face limitations in handling
关键词: Quaternions |  Vectors |  Image reconstruction |  Convolution |  Neural networks |  Face recognition |  Quantization (signal) |  Decoding |  Training |  Indexes

3. Anomaly detection scheme for lung CT images using vector quantized variational auto-encoder with support vector data description NSTL国家科技图书文献中心

Gao, Zhihui |  Nakayama, Ryohei... -  《Radiological physics and technology》 - 2025,18(1) - 17~26 - 共10页

摘要: introduced to VQ-VAE (Vector Quantized-Variational Auto |  this study, SVDD (Support Vector Data Description |  replaced by the embedding representations. The data of | This study aims to develop an anomaly | -detection scheme for lesions in CT images. Our database
关键词: Anomaly detection |  Lung CT images |  VQ-VAE |  SVDD

4. Trajectory Forecasting Through Low-Rank Adaptation of Discrete Latent Codes NSTL国家科技图书文献中心

Riccardo Benaglia |  Angelo Porrello... -  《Pattern Recognition,Part XVI》 -  International Conference on Pattern Recognition - 2025, - 236~251 - 共16页

摘要: leveraging Vector Quantized Variational Autoencoders (VQ |  tailored latent representations for each example. In a | Trajectory forecasting is crucial for video |  surveillance analytics, as it enables the anticipation of |  future movements for a set of agents, e.g., basketball
关键词: Trajectory forecasting |  Vector quantization

5. QSIM: A Quantum-inspired hierarchical semantic interaction model for text classification NSTL国家科技图书文献中心

Gao H. |  Zhang P.... -  《Neurocomputing》 - 2025,611(Jan.1) - 1.1~1.15 - 共15页

摘要: constructed quantized representations of language systems in |  the unit vector in the two-dimensional minimum | © 2024Semantic interaction modeling is a |  fundamental technology in natural language understanding |  that guides models to extract deep semantic
关键词: Attention mechanism |  Entanglement entropy |  Quantum language model |  Schmidt decomposition

6. Unsupervised EEG-Based Seizure Anomaly Detection with Denoising Diffusion Probabilistic Models NSTL国家科技图书文献中心

Wang, Jiale |  Sun, Mengxue... -  《International Journal of Neural Systems》 - 2024,34(9) - 共14页

摘要: spectrograms are encoded into vector-quantized |  representations, which enables powerful and efficient DDPM while | While many seizure detection methods have |  demonstrated great accuracy, their training necessitates a |  substantial volume of labeled data. To address this issue
关键词: Seizure detection |  denoising diffusion probabilistic models |  anomaly detection |  unsupervised learning |  vector-quantized representations

7. Patch-wise vector quantization for unsupervised medical anomaly detection NSTL国家科技图书文献中心

Kim T. |  Jeong I.... -  《Pattern recognition letters》 - 2024,184(Aug.) - 205~211 - 共7页

摘要: vector-quantized representations of normal “patches |  pathology detection method called Patch-wise Vector |  incorporating vector similarity metric. Furthermore, we |  collapse” problem of vector quantization by proposing top | ? 2024 Elsevier B.V.Radiography images
关键词: Anomaly detection |  Computer vision |  Medical imaging |  Representation learning

8. ε-ViLM : Efficient Video-Language Model via Masked Video Modeling with Semantic Vector-Quantized Tokenizer NSTL国家科技图书文献中心

Jacob Zhiyuan Fang |  Skyler Zheng... -  《2024 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops, [v.1]: WACVW 2024, Waikoloa, Hawaii, USA, 1-6 January 2024》 -  IEEE/CVF Conference on Applications of Computer Vision Workshops - 2024, - 529~540 - 共12页

摘要:, produced by the pre-trained vector-quantized tokenizer |  (MVM) schema, assisted with a semantic vector | -quantized tokenizer: In particular, our E-ViLM learns to |  representations from Video-Language corpus and generalize well | To build scalable models for the challenging
关键词: Visualization |  Computational modeling |  Architecture |  Semantics |  Buildings |  Focusing |  Question answering (information retrieval)

9. Robust unsupervised-learning based crack detection for stamped metal products NSTL国家科技图书文献中心

Zhang, Penghua |  Ryu, Hojun... -  《Journal of Manufacturing Systems》 - 2024,73 - 65~74 - 共10页

摘要: approach leverages the Vector Quantized-Variational | , latent features at different scales are quantized into |  discrete representations using a codebook. To learn the |  distribution of these discrete representations from non-crack |  highest dissimilarity. The edited representations are
关键词: Unsupervised learning |  Crack detection |  Reconstruction |  Stamped metal product |  VISION

10. Enhancing Image Generation by fusing Auto Encoder & Transformative Generation Approach NSTL国家科技图书文献中心

Eliza Femi Sherley S |  Sanjay T... -  《2024 International Conference on Electrical Electronics and Computing Technologies》 -  International Conference on Electrical Electronics and Computing Technologies - 2024, - 1~6 - 共6页

摘要: novel approach by fusing GPT with Vector Quantized |  derive discrete latent representations from input |  images. These latent representations, incorporating |  the latent space representations obtained from VQ | Generative Pre-trained Transformer (GPT
关键词: Adaptation models |  Visualization |  Image synthesis |  Image color analysis |  Computational modeling |  Gray-scale |  Transformers |  Data models |  Vectors |  Optimization
检索条件vector-quantized representations
  • 检索词扩展

NSTL主题词

  • NSTL学科导航