{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/temporal-interpolation-via-motion-field","title":"Temporal Interpolation via Motion Field Prediction","arxiv_id":"1804.04440","date":"2018-04-12","proceeding":null,"authors":["Lin Zhang","Neerav Karani","Christine Tanner","Ender Konukoglu"],"abstract":"Navigated 2D multi-slice dynamic Magnetic Resonance (MR) imaging enables high\ncontrast 4D MR imaging during free breathing and provides in-vivo observations\nfor treatment planning and guidance. Navigator slices are vital for\nretrospective stacking of 2D data slices in this method. However, they also\nprolong the acquisition sessions. Temporal interpolation of navigator slices an\nbe used to reduce the number of navigator acquisitions without degrading\nspecificity in stacking. In this work, we propose a convolutional neural\nnetwork (CNN) based method for temporal interpolation via motion field\nprediction. The proposed formulation incorporates the prior knowledge that a\nmotion field underlies changes in the image intensities over time. Previous\napproaches that interpolate directly in the intensity space are prone to\nproduce blurry images or even remove structures in the images. Our method\navoids such problems and faithfully preserves the information in the image.\nFurther, an important advantage of our formulation is that it provides an\nunsupervised estimation of bi-directional motion fields. We show that these\nmotion fields can be used to halve the number of registrations required during\n4D reconstruction, thus substantially reducing the reconstruction time.","url_abs":"http://arxiv.org/abs/1804.04440v1","url_pdf":"http://arxiv.org/pdf/1804.04440v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"temporal-interpolation-via-motion-field","repo_url":"https://github.com/linz94/mfin-cycle","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"4d-reconstruction","task_name":"4D reconstruction"},{"task_slug":"prediction","task_name":"Prediction"},{"task_slug":"specificity","task_name":"Specificity"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}