NSTL国家科技图书文献中心

1. Neural Image Compression with Regional Decoding NSTL国家科技图书文献中心

YILI JIN | JIAHAO LI ... - 《ACM transactions on multimedia computing communications and applications》 - 2025,21(3) - 73.1～73.18 - 共18页

摘要：. However, existing neural* image compression methods lack* | growing need for a function in image* compression named* | application of our method to 360 image compression | regional decoding. This function lets an image* be encoded* | that, during the compression* process, local*

关键词： Image compression | Neural networks | Regional decoding

2. Rate-Distortion-Cognition Controllable Versatile Neural Image Compression NSTL国家科技图书文献中心

Jinming Liu | Ruoyu Feng ... - 《Computer Vision - ECCV 2024,Part LVI》 - European Conference on Computer Vision - 2025, - 329～348 - 共20页

摘要： learning-based techniques for image* compression and* | versatile image* compression, which method allows the users* | Recently, the field of Image* Coding for* | to adjust the bitrate (i.e., Rate), image | accuracy (i.e., Cognition) with a single neural* model*

关键词： Neural image compression | Image compression for machine | Variable-bitrate compression | Cognition-distortion trade-off

3. BaSIC: BayesNet Structure Learning for Computational Scalable Neural Image Compression NSTL国家科技图书文献中心

Yufeng Zhang | Hang Yu ... - 《Computer Vision - ECCV 2024,Part XXV》 - European Conference on Computer Vision - 2025, - 269～285 - 共17页

摘要： over traditional codecs, Neural Image Compression | learning for computational Scalable neural Image | Compression), a comprehensive, computationally scalable | for controlling both neural* network backbones and* | control but also maintains competitive compression

关键词： Image compression | Computational scalable compression | BayesNet structure learning | Autoregressive model

4. Stable successive Neural Image Compression via coherent demodulation-based transformation NSTL国家科技图书文献中心

Bao, Youneng | Tan, Wen ... - 《Signal Processing》 - 2025,227 - 共13页

摘要：Neural Image Compression (NIC) has made | iterative re-compression cycles, which can degrade image | improve the re-compression* stability but also preserves* | significant strides in recent years. However, the existing | NIC methods demonstrate instability issues during

关键词： Deep learning | Communication system | Image compression | Idempotent

5. HNRC: Lightweight Image Compression with Hybrid Neural Representation NSTL国家科技图书文献中心

Xinyuan Cheng | Dongdong Zhang ... - 《Pattern Recognition,Part XXXI》 - International Conference on Pattern Recognition - 2025, - 404～418 - 共15页

摘要： image compression method with hybrid neural | Recently, image* compression methods based on* | and forms INR-based image* compression method* | Neural Representations (INR) represents images as | neural networks that map coordinates to signal values

关键词： Image compression | Implicit neural representation | Hybrid neural representation

6. Hyperspectral Image Compression Using Sampling and Implicit Neural Representations NSTL国家科技图书文献中心

Shima Rezasoltani | Faisal Z. Qureshi - 《IEEE Transactions on Geoscience and Remote Sensing》 - 2025,63 - 1～12 - 共12页

摘要： hyperspectral image compression using implicit neural | hyperspectral image* contains an order of magnitude more* | information than a similar-sized RGB color image | hyperspectral image. This representation, thus, acts as a | compressed encoding of this image, and the original image

关键词： Image coding | Hyperspectral imaging | Transform coding | Neural networks | Image reconstruction | Principal component analysis | Color | Gray-scale | Three-dimensional displays | Tensors

7. Neural Graphics Texture Compression Supporting Random Access NSTL国家科技图书文献中心

Farzad Farhadzadeh | Qiqi Hou ... - 《Computer Vision - ECCV 2024,Part XXXVII》 - European Conference on Computer Vision - 2025, - 412～429 - 共18页

摘要： its compression. Meanwhile Neural Image Compression | adapted to neural* texture compression. First, texture* | compression requires on-demand and real-time decoding with | set compression* that integrates traditional GPU* | than conventional texture compression, and

关键词： Graphics texture | Neural implicit representation | Neural image compression | Random access

8. Scalable and Resolution Data Analysis of Image and Video Compression using DL-CNNS Neural Network NSTL国家科技图书文献中心

A. J. Ajal | S. Anbu Karuppusamy - 《Circuits, systems, and signal processing》 - 2025,44(3) - 1845～1864 - 共20页

摘要： CNN based on DL for video and image* compression to* | cost. The results of the image* compression experiment* | processing them. High-quality image* and video footage is* | compression. The DL-based CNN model's output vector | % decrease in RD cost during a video compression* experiment*

关键词： Convolution Neural Networks | Deep learning | Video and Image processing | Compression | JPEG

9. Unsupervised Degradation Representation Learning for Unpaired Restoration of Images and Point Clouds NSTL国家科技图书文献中心

Longguang Wang | Yulan Guo ... - 《IEEE Transactions on Pattern Analysis and Machine Intelligence》 - 2025,47(1) - 1～18 - 共18页

摘要： UnIRnet and UnPRnet for unpaired image* and point cloud* | information. Experiments on unpaired image* and point cloud* | Restoration tasks in low-level vision aim to | restore high-quality (HQ) data from their low-quality | (LQ) observations. To circumvents the difficulty of

关键词： Degradation | Image restoration | Point cloud compression | Training | Data mining | Representation learning | Neural networks | Image denoising | Noise | Convolutional neural networks

10. Asymmetric Learned Image Compression Using Fast Residual Channel Attention NSTL国家科技图书文献中心

Yusong Hu | Cheolkon Jung ... - 《Pattern Recognition,Part XXVI》 - International Conference on Pattern Recognition - 2025, - 156～170 - 共15页

摘要： been made in deep learning-based image compression | paper, we propose asymmetric learned image compression | asymmetric image* compression network to effectively* | lost by compression, resulting in image* quality* | learned image* compression in terms of PSNR, MS-SSIM and*

摘要：. However, existing neural* image compression methods lack* | growing need for a function in image* compression named* | application of our method to 360 image compression | regional decoding. This function lets an image* be encoded* | that, during the compression* process, local*

关键词： Image compression | Neural networks | Regional decoding

摘要： learning-based techniques for image* compression and* | versatile image* compression, which method allows the users* | Recently, the field of Image* Coding for* | to adjust the bitrate (i.e., Rate), image | accuracy (i.e., Cognition) with a single neural* model*

关键词： Neural image compression | Image compression for machine | Variable-bitrate compression | Cognition-distortion trade-off

摘要： over traditional codecs, Neural Image Compression | learning for computational Scalable neural Image | Compression), a comprehensive, computationally scalable | for controlling both neural* network backbones and* | control but also maintains competitive compression

关键词： Image compression | Computational scalable compression | BayesNet structure learning | Autoregressive model

摘要：Neural Image Compression (NIC) has made | iterative re-compression cycles, which can degrade image | improve the re-compression* stability but also preserves* | significant strides in recent years. However, the existing | NIC methods demonstrate instability issues during

关键词： Deep learning | Communication system | Image compression | Idempotent

摘要： image compression method with hybrid neural | Recently, image* compression methods based on* | and forms INR-based image* compression method* | Neural Representations (INR) represents images as | neural networks that map coordinates to signal values

关键词： Image compression | Implicit neural representation | Hybrid neural representation

摘要： hyperspectral image compression using implicit neural | hyperspectral image* contains an order of magnitude more* | information than a similar-sized RGB color image | hyperspectral image. This representation, thus, acts as a | compressed encoding of this image, and the original image

关键词： Image coding | Hyperspectral imaging | Transform coding | Neural networks | Image reconstruction | Principal component analysis | Color | Gray-scale | Three-dimensional displays | Tensors

摘要： its compression. Meanwhile Neural Image Compression | adapted to neural* texture compression. First, texture* | compression requires on-demand and real-time decoding with | set compression* that integrates traditional GPU* | than conventional texture compression, and

关键词： Graphics texture | Neural implicit representation | Neural image compression | Random access

摘要： CNN based on DL for video and image* compression to* | cost. The results of the image* compression experiment* | processing them. High-quality image* and video footage is* | compression. The DL-based CNN model's output vector | % decrease in RD cost during a video compression* experiment*

关键词： Convolution Neural Networks | Deep learning | Video and Image processing | Compression | JPEG

摘要： UnIRnet and UnPRnet for unpaired image* and point cloud* | information. Experiments on unpaired image* and point cloud* | Restoration tasks in low-level vision aim to | restore high-quality (HQ) data from their low-quality | (LQ) observations. To circumvents the difficulty of

关键词： Degradation | Image restoration | Point cloud compression | Training | Data mining | Representation learning | Neural networks | Image denoising | Noise | Convolutional neural networks

摘要： been made in deep learning-based image compression | paper, we propose asymmetric learned image compression | asymmetric image* compression network to effectively* | lost by compression, resulting in image* quality* | learned image* compression in terms of PSNR, MS-SSIM and*

关键词： Image compression | Convolutional neural network | Depthwise separable convolution | Fast residual channel attention | Variational auto-encoder

4008-161-200 800-990-8900

国家科技图书文献中心

摘要：. However, existing neural image compression methods lack | growing need for a function in image compression named | application of our method to 360 image compression | regional decoding. This function lets an image be encoded | that, during the compression process, local

关键词： Image compression | Neural networks | Regional decoding

摘要： learning-based techniques for image compression and | versatile image compression, which method allows the users | Recently, the field of Image Coding for | to adjust the bitrate (i.e., Rate), image | accuracy (i.e., Cognition) with a single neural model

关键词： Neural image compression | Image compression for machine | Variable-bitrate compression | Cognition-distortion trade-off

摘要： over traditional codecs, Neural Image Compression | learning for computational Scalable neural Image | Compression), a comprehensive, computationally scalable | for controlling both neural network backbones and | control but also maintains competitive compression

关键词： Image compression | Computational scalable compression | BayesNet structure learning | Autoregressive model

摘要：Neural Image Compression (NIC) has made | iterative re-compression cycles, which can degrade image | improve the re-compression stability but also preserves | significant strides in recent years. However, the existing | NIC methods demonstrate instability issues during

关键词： Deep learning | Communication system | Image compression | Idempotent

摘要： image compression method with hybrid neural | Recently, image compression methods based on | and forms INR-based image compression method | Neural Representations (INR) represents images as | neural networks that map coordinates to signal values

关键词： Image compression | Implicit neural representation | Hybrid neural representation

摘要： hyperspectral image compression using implicit neural | hyperspectral image contains an order of magnitude more | information than a similar-sized RGB color image | hyperspectral image. This representation, thus, acts as a | compressed encoding of this image, and the original image

关键词： Image coding | Hyperspectral imaging | Transform coding | Neural networks | Image reconstruction | Principal component analysis | Color | Gray-scale | Three-dimensional displays | Tensors

摘要： its compression. Meanwhile Neural Image Compression | adapted to neural texture compression. First, texture | compression requires on-demand and real-time decoding with | set compression that integrates traditional GPU | than conventional texture compression, and

关键词： Graphics texture | Neural implicit representation | Neural image compression | Random access

摘要： CNN based on DL for video and image compression to | cost. The results of the image compression experiment | processing them. High-quality image and video footage is | compression. The DL-based CNN model's output vector | % decrease in RD cost during a video compression experiment

关键词： Convolution Neural Networks | Deep learning | Video and Image processing | Compression | JPEG

摘要： UnIRnet and UnPRnet for unpaired image and point cloud | information. Experiments on unpaired image and point cloud | Restoration tasks in low-level vision aim to | restore high-quality (HQ) data from their low-quality | (LQ) observations. To circumvents the difficulty of

关键词： Degradation | Image restoration | Point cloud compression | Training | Data mining | Representation learning | Neural networks | Image denoising | Noise | Convolutional neural networks

摘要： been made in deep learning-based image compression | paper, we propose asymmetric learned image compression | asymmetric image compression network to effectively | lost by compression, resulting in image quality | learned image compression in terms of PSNR, MS-SSIM and

关键词： Image compression | Convolutional neural network | Depthwise separable convolution | Fast residual channel attention | Variational auto-encoder

4008-161-200 800-990-8900

国家科技图书文献中心

摘要：. However, existing neural* image compression methods lack* | growing need for a function in image* compression named* | application of our method to 360 image compression | regional decoding. This function lets an image* be encoded* | that, during the compression* process, local*

摘要： learning-based techniques for image* compression and* | versatile image* compression, which method allows the users* | Recently, the field of Image* Coding for* | to adjust the bitrate (i.e., Rate), image | accuracy (i.e., Cognition) with a single neural* model*

摘要： over traditional codecs, Neural Image Compression | learning for computational Scalable neural Image | Compression), a comprehensive, computationally scalable | for controlling both neural* network backbones and* | control but also maintains competitive compression

摘要：Neural Image Compression (NIC) has made | iterative re-compression cycles, which can degrade image | improve the re-compression* stability but also preserves* | significant strides in recent years. However, the existing | NIC methods demonstrate instability issues during

摘要： image compression method with hybrid neural | Recently, image* compression methods based on* | and forms INR-based image* compression method* | Neural Representations (INR) represents images as | neural networks that map coordinates to signal values

摘要： hyperspectral image compression using implicit neural | hyperspectral image* contains an order of magnitude more* | information than a similar-sized RGB color image | hyperspectral image. This representation, thus, acts as a | compressed encoding of this image, and the original image

摘要： its compression. Meanwhile Neural Image Compression | adapted to neural* texture compression. First, texture* | compression requires on-demand and real-time decoding with | set compression* that integrates traditional GPU* | than conventional texture compression, and

摘要： CNN based on DL for video and image* compression to* | cost. The results of the image* compression experiment* | processing them. High-quality image* and video footage is* | compression. The DL-based CNN model's output vector | % decrease in RD cost during a video compression* experiment*

摘要： UnIRnet and UnPRnet for unpaired image* and point cloud* | information. Experiments on unpaired image* and point cloud* | Restoration tasks in low-level vision aim to | restore high-quality (HQ) data from their low-quality | (LQ) observations. To circumvents the difficulty of

摘要： been made in deep learning-based image compression | paper, we propose asymmetric learned image compression | asymmetric image* compression network to effectively* | lost by compression, resulting in image* quality* | learned image* compression in terms of PSNR, MS-SSIM and*