SYSTEM FOR GENERATING IMAGE AND NON-TRANSITORY COMPUTER-READABLE MEDIUM

To efficiently learn specific parts.SOLUTION: A system configured to generate a transformed image on the basis of an input image input includes a learning model whose parameters are adjusted so as to suppress an error between the input image and a second image transformed by the input of the input i...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	TOYODA YASUTAKA, ISHIKAWA MASAYOSHI, OUCHI MASAKI
Format:	Patent
Sprache:	eng ; jpn
Schlagworte:	BASIC ELECTRIC ELEMENTS CALCULATING COMPUTING COUNTING ELECTRIC DISCHARGE TUBES OR DISCHARGE LAMPS ELECTRICITY IMAGE DATA PROCESSING OR GENERATION, IN GENERAL PHYSICS
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Beschreibung
Zusammenfassung:	To efficiently learn specific parts.SOLUTION: A system configured to generate a transformed image on the basis of an input image input includes a learning model whose parameters are adjusted so as to suppress an error between the input image and a second image transformed by the input of the input image, and the learning model performs different learning on at least a first region in the image and a second region different from the first region.SELECTED DRAWING: Figure 11 【課題】効率良く、特定部位の学習を行う。【解決手段】入力画像の入力に基づいて、変換画像を生成するように構成されたシステムであって、前記入力画像の入力によって変換された第２の画像との間の誤差を抑制するように、パラメータが調整された学習モデルを含み、当該学習モデルは、少なくとも画像内の第１の領域と、当該第１の領域とは異なる第２の領域との間で異なる学習が施されているシステムを提案する。【選択図】図１１