Rotationally Equivariant Super-Resolution of Velocity Fields in Two-Dimensional Fluids Using Convolutional Neural Networks
원문 초록
This paper investigates the super-resolution (SR) of velocity fields in two-dimensional fluids from the viewpoint of rotational equivariance. SR refers to techniques that estimate high-resolution images from those in low resolution and has lately been applied in fluid mechanics. The rotational equivariance of SR models is defined as the property in which the super-resolved velocity field is rotated according to a rotation of the input, which leads to the inference covariant to the orientation of fluid systems. Generally, the covariance in physics is related to symmetries. To clarify a relationship to symmetries, the rotational consistency of datasets for SR is newly introduced as the invariance of pairs of low- and high-resolution velocity fields with respect to rotation. This consistency is sufficient and necessary for SR models to acquire rotational equivariance from large datasets with supervised learning. Such a large dataset is not required when rotational equivariance is imposed on SR models through weight sharing of convolution kernels as prior knowledge. Even if a fluid system has rotational symmetry, this symmetry may not carry over to a velocity dataset, which is not rotationally consistent. This inconsistency can occur when the rotation does not commute with the generation of low-resolution velocity fields. These theoretical suggestions are supported by the results from numerical experiments, where two existing convolutional neural networks (CNNs) are converted into rotationally equivariant CNNs and the inferences of the four CNNs are compared after the supervised training.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
Unsupervised Image Super-Resolution Reconstruction Based on Real-World Degradation Patterns
그리드는 종종 밀집 신호 압축에서 암시적 신경 표현(Implicit Neural Representations, INRs)보다 우수합니다
Grids Often Outperform Implicit Neural Representations at Compressing Dense Signals
암시적 신경 표현(Implicit Neural Representations, INRs)은 최근 인상적인 결과를 보여주었지만, 그 기본 용량, 암시적 편향, 스케일링 동작은 여전히 잘 이해되지 않고 있습니다. 우리는 다양한 유효 대역폭을 가진 2D 및 3D 실제 및 합성 신호와 토모그래피, 초해상도, 노이즈 제거를 포함한 과적합(overfitting) 및 일반화 작업 전반에 걸쳐 다양한 INRs의 성능을 조사합니다. 모델 크기와 신호 유형 및 대역폭에 따라 성능을 계층화하여, 서로 다른 INR 및 그리드 표현이 용량을 어떻게 할당하는지를 밝혀냅니다. 밀집 신호를 다루는 많은 작업에서, 보간이 적용된 간단한 정규화된 그리드는 동일한 파라미터 수를 가진 어떤 INR보다 더 빠르게 학습되며 더 높거나 대등한 품질로 도달함을 발견했습니다. 또한, 형태 윤곽과 같은 이진 신호 적합 작업과 같이 INRs가 그리드보다 우수한 성능을 보이는 제한적인 설정도 발견했습니다. 이는 향후 INRs의 개발과 활용을 가장 유리한 응용 분야로 유도하는 데 도움이 될 것입니다.
IM-LUT: Interpolation Mixing Look-Up Tables for Image Super-Resolution
PanoDiff-SR: Synthesizing Dental Panoramic Radiographs using Diffusion and Super-resolution
HNOSeg-XS: Extremely Small Hartley Neural Operator for Efficient and Resolution-Robust 3D Image Segmentation
CycleVAR: Repurposing Autoregressive Model for Unsupervised One-Step Image Translation
토론
이 논문과 연결된 포럼 글이 아직 없습니다.