{"url":"/method/z-pnn","slug":"z-pnn","name":"Z-PNN","full_name":"Pansharpening by convolutional neural networks in the full resolution framework","full_name_withheld":false,"description_markdown":"In recent years, there has been a growing interest on deep learning-based pansharpening.\r\nResearch has mainly focused on architectures.\r\nHowever, lacking a ground truth, model training is also a major issue.\r\nA popular approach is to train networks in a reduced resolution domain, using the original data as ground truths.\r\nThe trained networks are then used on full resolution data, relying on an implicit scale invariance hypothesis.\r\nResults are generally good at reduced resolution, but more questionable at full resolution.\r\n\r\nHere, we propose a full-resolution training framework for deep learning-based pansharpening.\r\nTraining takes place in the high resolution domain, relying only on the original data, with no loss of information.\r\nTo ensure spectral and spatial fidelity, suitable losses are defined,\r\nwhich force the pansharpened output to be consistent with the available panchromatic and multispectral input.\r\nExperiments carried out on WorldView-3, WorldView-2, and GeoEye-1 images show that methods trained with the proposed framework\r\nguarantee an excellent performance in terms of both full-resolution numerical indexes and visual quality.\r\nThe framework is fully general, and can be used to train and fine-tune any deep learning-based pansharpening network.","description_state":"present","introduced_year":null,"introduced_by":{"title":"Pansharpening by convolutional neural networks in the full resolution framework","paper":"/paper/pansharpening-by-convolutional-neural","first_author":"Matteo Ciotola","n_authors":5,"url_abs":null,"archive_paper_url":"https://paperswithcode.com/paper/pansharpening-by-convolutional-neural"},"source":{"url":"https://arxiv.org/abs/2111.08334v3","title":"Pansharpening by convolutional neural networks in the full resolution framework","url_on_a_paper_host":true},"code_snippet_url":null,"code_snippet_url_on_a_code_host":false,"categories":[{"area":"Computer Vision","area_id":"computer-vision","collection":"Convolutional Neural Networks","url":"/methods/category/convolutional-neural-networks","pwc_aliases":[]}],"n_papers_tagged":1,"archive_num_papers":1,"papers_newest_first":[{"paper":"/paper/pansharpening-by-convolutional-neural","title":"Pansharpening by convolutional neural networks in the full resolution framework","date":"2021-11-16","arxiv_id":"2111.08334","n_code_links":2,"syntology":null}],"papers_shown":1,"tasks":[{"task":"/task/image-super-resolution","name":"Image Super-Resolution","papers":1},{"task":"/task/pansharpening","name":"Pansharpening","papers":1},{"task":"/task/super-resolution","name":"Super-Resolution","papers":1},{"task":"/task/satellite-image-super-resolution","name":"satellite image super-resolution","papers":1}],"tasks_shown":4,"n_tasks":4,"usage_by_year":[{"year":"2021","papers":1}],"row_source":"methods_table","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/z-pnn"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}