NSTL国家科技图书文献中心

1. A review of optical flow velocimetry in fluid mechanics NSTL国家科技图书文献中心

Jassal, Gauresh Raj | Schmidt, Bryan E. - 《Measurement Science & Technology》 - 2025,36(3) - 1～20 - 共20页 - 被引量：1

摘要：Optical flow methods have been developed over | techniques. Numerous optical* flow velocimetry (OFV* | ) algorithms have been devised to solve the ill-posed optical | flow problem, with various physics-inspired | the past two decades for application to particle

关键词： optical flow | PIV | velocimetry

2. Learning 360° Optical Flow Using Tangent Images and Transformer NSTL国家科技图书文献中心

Yanjie Ma | Cheng Han ... - 《Pattern Recognition and Computer Vision,Part III》 - Chinese Conference on Pattern Recognition and Computer Vision - 2025, - 144～157 - 共14页

摘要：Optical flow estimation has always been a | perspective-based methods for optical* flow estimation fail* | ° panoramic images, making optical* flow estimation for* | fundamental task in computer vision. Due to the ultra-wide | field of view (FoV) of panoramic cameras, traditional

关键词： Optical flow | Panoramic image | Distortion | Transformer

3. SDformerFlow: Spiking Neural Network Transformer for Event-based Optical Flow NSTL国家科技图书文献中心

Yi Tian | Juan Andrade-Cetto - 《Pattern Recognition,Part XV》 - International Conference on Pattern Recognition - 2025, - 475～491 - 共17页

摘要： fast and robust optical* flow estimation: STTFlowNet* | for dense optical* flow estimation. We conduct end-to* | learning on the DSEC-flow* Dataset. Our results indicate* | Event cameras produce asynchronous and sparse | event streams capturing changes in light intensity

关键词： Spiking neural network | Event camera | Transformer | Optical flow

4. Fast recurrent field transforms for optical flow on edge GPUs NSTL国家科技图书文献中心

Shi, Ke | Miao, Yinxiao ... - 《Measurement Science & Technology》 - 2025,36(3) - 1～11 - 共11页

摘要：Optical flow reflects motion information in | computational demands make dense optical* flow estimation* | by introducing EdgeFlow, an optical* flow network* | gated core for optical* flow update, capable of* | videos, serving as a crucial visual cue. However, high

关键词： optical flow estimation | deep learning | edge devices | model lightweight | real-time processing

5. EdgeFlowNet: 100FPS@1W Dense Optical Flow for Tiny Mobile Robots NSTL国家科技图书文献中心

Sai Ramana Kiran Pin... | Rishabh Singh ... - 《IEEE robotics and automation letters》 - 2025,10(1) - 128～135 - 共8页

摘要：Optical flow estimation is a critical task for | . However, optical* flow estimation on tiny robots is* | -speed, low-latency dense optical* flow approach for* | tiny mobile robotics to enable safe and accurate | navigation, obstacle avoidance, and other functionalities

关键词： Optical flow | Accuracy | Robot sensing systems | Robots | Navigation | Image edge detection | Estimation | Quadrotors | Computer architecture | Collision avoidance

6. LayeredFlow: A Real-World Benchmark for Non-Lambertian Multi-layer Optical Flow NSTL国家科技图书文献中心

Hongyu Wen | Erich Liang ... - 《Computer Vision - ECCV 2024,Part II》 - European Conference on Computer Vision - 2025, - 477～495 - 共19页

摘要： for optical* flow of non-Lambertian objects. Compared* | quality optical* flow and stereo pairs taken over 185* | multi-layer optical* flow. To provide training data for* | predict multilayer optical* flow, while fine-tuning* | existing optical* flow methods on the dataset notably*

关键词： Non-Lambertian object | Optical flow | Benchmark | Dataset

7. MREIFlow: Unsupervised dense and time-continuous optical flow estimation from and event data NSTL国家科技图书文献中心

Hu, Jianlang | Guo, Chi ... - 《Information Fusion》 - 2025,113 - 共14页

摘要： time-continuous optical* flow. Methods: We propose a* | multi-scale optical* flow recurrent estimation network* | full-resolution optical* flow iteratively in a coarse* | and time-continuous optical* flow estimation* | -continuous optical* flow estimation, and can be trained*

关键词： Optical flow | Image | Event camera | Unsupervised learning

8. A Variational Formulation of the Euler Deconvolution Method for Gravity Data Inspired by Optical Flow NSTL国家科技图书文献中心

Leon, Jose Antonio R... | Becerra, Emilia Freg...... - 《Pure and Applied Geophysics》 - 2025,182(5) - 2001～2022 - 共22页

摘要： the optical* flow within a sequence of images. A* | popular method for estimating the optical* flow is the* | inversion and optical* flow are different phenomena, the* | formulation for the optical* flow problem that improves the* | The Euler deconvolution method has been widely

关键词： Geophysical data inversion | euler deconvolution | optical flow | variational formulation

9. Sensitivity of wavelet-based optical flow velocimetry (wOFV) to common experimental error sources NSTL国家科技图书文献中心

Schmidt, Bryan E. | Page, Wayne E.... - 《Measurement Science & Technology》 - 2025,36(1) - 1～12 - 共12页 - 被引量：1

摘要： accuracy of a wavelet-based optical* flow velocimetry* | improvement is consistent between wOFV and PIV. Flow | The influence of several potential error | sources and non-ideal experimental effects on the | (wOFV) method when applied to tracer particle images

关键词： fluid velocimetry | optical flow | PIV

10. Learned Optical Flow-Based Defect Detection With Coarse-to-Fine Nonrigid Image Alignment NSTL国家科技图书文献中心

Deng, Lei | Meng, Xiangze ... - 《IEEE sensors journal》 - 2025,25(5) - 8684～8696 - 共13页

摘要： alignment framework with learned optical* flow for PCB* | optical flow estimation module with a multiscale | optical flow field. Finally, we employ the Mahalanobis | generalization properties of optical* flow. Extensive* | optical inspection (AOI) systems equipped with machine

关键词： Optical flow | Defect detection | Deformation | Printed circuits | Image edge detection | Feature extraction | Vibrations | Image reconstruction | Transformers | Symmetric matrices...

摘要：Optical flow methods have been developed over | techniques. Numerous optical flow velocimetry (OFV | ) algorithms have been devised to solve the ill-posed optical | flow problem, with various physics-inspired | the past two decades for application to particle

关键词： optical flow | PIV | velocimetry

摘要：Optical flow estimation has always been a | perspective-based methods for optical flow estimation fail | ° panoramic images, making optical flow estimation for | fundamental task in computer vision. Due to the ultra-wide | field of view (FoV) of panoramic cameras, traditional

关键词： Optical flow | Panoramic image | Distortion | Transformer

摘要： fast and robust optical flow estimation: STTFlowNet | for dense optical flow estimation. We conduct end-to | learning on the DSEC-flow Dataset. Our results indicate | Event cameras produce asynchronous and sparse | event streams capturing changes in light intensity

关键词： Spiking neural network | Event camera | Transformer | Optical flow

摘要：Optical flow reflects motion information in | computational demands make dense optical flow estimation | by introducing EdgeFlow, an optical flow network | gated core for optical flow update, capable of | videos, serving as a crucial visual cue. However, high

关键词： optical flow estimation | deep learning | edge devices | model lightweight | real-time processing

摘要：Optical flow estimation is a critical task for | . However, optical flow estimation on tiny robots is | -speed, low-latency dense optical flow approach for | tiny mobile robotics to enable safe and accurate | navigation, obstacle avoidance, and other functionalities

关键词： Optical flow | Accuracy | Robot sensing systems | Robots | Navigation | Image edge detection | Estimation | Quadrotors | Computer architecture | Collision avoidance

摘要： for optical flow of non-Lambertian objects. Compared | quality optical flow and stereo pairs taken over 185 | multi-layer optical flow. To provide training data for | predict multilayer optical flow, while fine-tuning | existing optical flow methods on the dataset notably

关键词： Non-Lambertian object | Optical flow | Benchmark | Dataset

摘要： time-continuous optical flow. Methods: We propose a | multi-scale optical flow recurrent estimation network | full-resolution optical flow iteratively in a coarse | and time-continuous optical flow estimation | -continuous optical flow estimation, and can be trained

关键词： Optical flow | Image | Event camera | Unsupervised learning

摘要： the optical flow within a sequence of images. A | popular method for estimating the optical flow is the | inversion and optical flow are different phenomena, the | formulation for the optical flow problem that improves the | The Euler deconvolution method has been widely

关键词： Geophysical data inversion | euler deconvolution | optical flow | variational formulation

摘要： accuracy of a wavelet-based optical flow velocimetry | improvement is consistent between wOFV and PIV. Flow | The influence of several potential error | sources and non-ideal experimental effects on the | (wOFV) method when applied to tracer particle images

关键词： fluid velocimetry | optical flow | PIV

摘要： alignment framework with learned optical flow for PCB | optical flow estimation module with a multiscale | optical flow field. Finally, we employ the Mahalanobis | generalization properties of optical flow. Extensive | optical inspection (AOI) systems equipped with machine

关键词： Optical flow | Defect detection | Deformation | Printed circuits | Image edge detection | Feature extraction | Vibrations | Image reconstruction | Transformers | Symmetric matrices...

4008-161-200 800-990-8900

国家科技图书文献中心

摘要：Optical flow methods have been developed over | techniques. Numerous optical* flow velocimetry (OFV* | ) algorithms have been devised to solve the ill-posed optical | flow problem, with various physics-inspired | the past two decades for application to particle

摘要：Optical flow estimation has always been a | perspective-based methods for optical* flow estimation fail* | ° panoramic images, making optical* flow estimation for* | fundamental task in computer vision. Due to the ultra-wide | field of view (FoV) of panoramic cameras, traditional

摘要： fast and robust optical* flow estimation: STTFlowNet* | for dense optical* flow estimation. We conduct end-to* | learning on the DSEC-flow* Dataset. Our results indicate* | Event cameras produce asynchronous and sparse | event streams capturing changes in light intensity

摘要：Optical flow reflects motion information in | computational demands make dense optical* flow estimation* | by introducing EdgeFlow, an optical* flow network* | gated core for optical* flow update, capable of* | videos, serving as a crucial visual cue. However, high

摘要：Optical flow estimation is a critical task for | . However, optical* flow estimation on tiny robots is* | -speed, low-latency dense optical* flow approach for* | tiny mobile robotics to enable safe and accurate | navigation, obstacle avoidance, and other functionalities

摘要： for optical* flow of non-Lambertian objects. Compared* | quality optical* flow and stereo pairs taken over 185* | multi-layer optical* flow. To provide training data for* | predict multilayer optical* flow, while fine-tuning* | existing optical* flow methods on the dataset notably*

摘要： time-continuous optical* flow. Methods: We propose a* | multi-scale optical* flow recurrent estimation network* | full-resolution optical* flow iteratively in a coarse* | and time-continuous optical* flow estimation* | -continuous optical* flow estimation, and can be trained*

摘要： the optical* flow within a sequence of images. A* | popular method for estimating the optical* flow is the* | inversion and optical* flow are different phenomena, the* | formulation for the optical* flow problem that improves the* | The Euler deconvolution method has been widely

摘要： accuracy of a wavelet-based optical* flow velocimetry* | improvement is consistent between wOFV and PIV. Flow | The influence of several potential error | sources and non-ideal experimental effects on the | (wOFV) method when applied to tracer particle images

摘要： alignment framework with learned optical* flow for PCB* | optical flow estimation module with a multiscale | optical flow field. Finally, we employ the Mahalanobis | generalization properties of optical* flow. Extensive* | optical inspection (AOI) systems equipped with machine