Content-aware frame interpolation (CAFI): Deep Learning-based temporal super-resolution for fast bioimaging
Poster presented as part of the Crick BioImage Analysis Symposium 2023.
Observing fast cellular dynamics is challenging because of phototoxicity and photobleaching. We implement two content-aware frame interpolation networks (CAFI): Zooming SlowMo (ZS) and Depth-Aware Video Frame Interpolation (DAIN). ZS and DAIN predict images between image pairs, in Time and Z dimensions, better than standard interpolation. We benchmark CAFI on 12 datasets from 4 microscopy modalities (point-scanning confocal, spinning-disk confocal, lattice light sheet and confocal brightfield microscopy. CAFI improves tracking of simulated and real objects, and segmentation of labelled nuclei. CAFI allows for reduced exposure: go faster and reduce phototoxicity or get more images for same exposure.
Permission has been given by authors to upload to Crick Figshare. Copyright remains with the original authors.