130535 20240209144727.0 doi 10.1109/ICCVW60793.2023.00124 sideral 136545 ART-2023-136545 eng Berriel Martins, Tomás Universidad de Zaragoza (orcid)0000-0002-2555-374X Ray-Patch: an efficient querying for light field transformers 2023 Access copy available to the general public Unrestricted In this paper we propose the Ray-Patch querying, a novel model to efficiently query transformers to decode implicit representations into target views. Our Ray-Patch decoding reduces the computational footprint and increases inference speed up to one order of magnitude compared to previous models, without losing global attention, and hence maintaining specific task metrics. The key idea of our novel querying is to split the target image into a set of patches, then querying the transformer for each patch to extract a set of feature vectors, which are finally decoded into the target image using convolutional layers. Our experimental results quantify the effectiveness of our method, specifically the notable boost in rendering speed for the same task metrics. info:eu-repo/semantics/openAccess All rights reserved http://www.europeana.eu/rights/rr-f/ info:eu-repo/semantics/conferenceObject info:eu-repo/semantics/acceptedVersion Civera Sancho, Javier Universidad de Zaragoza (orcid)0000-0003-1368-1151 5007 520 Universidad de Zaragoza Dpto. Informát.Ingenie.Sistms. Área Ingen.Sistemas y Automát. (2023), 1150 - 1155 IEEE International Conference on Computer Vision workshops 2473-9936 1173654 http://zaguan.unizar.es/record/130535/files/texto_completo.pdf Postprint 2572610 http://zaguan.unizar.es/record/130535/files/texto_completo.jpg?subformat=icon icon Postprint oai:zaguan.unizar.es:130535 articulos driver 2024-02-09-14:46:13 ARTICLE