TSR-VFD: Generating temporal super-resolution for unsteady vector field data

We present TSR-VFD, a novel deep learning solution that recovers temporal super-resolution (TSR) of three-dimensional vector field data (VFD) for unsteady flow. In scientific visualization, TSR-VFD is the first work leverages neural nets to interpolate intermediate fields from temporally sparsely sampled fields. The core lies in using two networks: InterpolationNet and MaskNet, process components different scales as input jointly output synthesized To demonstrate our approach’s effectiveness, we report qualitative quantitative results with several sets compare against interpolation linear (LERP), generative adversarial network (GAN), recurrent (RNN). addition, lossy compression (LC) scheme. Finally, conduct comprehensive study evaluate critical parameter settings designs. • A framework proposed interpolation. Quantitative scores are best among start-of-the-art approaches. Rendering qualities both pathlines streamlines closest ground truth.