INFORMATION PROCESSING SYSTEM, AND INFORMATION PROCESSING METHOD

To efficiently extract an object region from a captured image.SOLUTION: An information processing system: stores a machine learning model obtained by learning data having a captured image, and region descriptive data which is data obtained by expressing an object region which is a region having a sp...

Full description

Saved in:
Bibliographic Details
Main Authors KAGA YOSUKE, FUJIO MASAKAZU, TAKAHASHI KENTA, NAKAMURA WATARU
Format Patent
LanguageEnglish
Japanese
Published 25.02.2021
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:To efficiently extract an object region from a captured image.SOLUTION: An information processing system: stores a machine learning model obtained by learning data having a captured image, and region descriptive data which is data obtained by expressing an object region which is a region having a specific object in the captured image with a one-dimensional numerical value sequence, associated with one another, as teacher data; and estimates the region descriptive data by inputting the captured image to the machine learning model. A background separation image is generated from the captured image, and the region descriptive data is generated based on the generated background separation image. The information processing system estimates the region descriptive data by inputting the captured image to the machine learning model, and restores the background separation image based on the region descriptive data.SELECTED DRAWING: Figure 1 【課題】撮影画像から対象領域を効率よく抽出する。【解決手段】情報処理システムは、撮影画像と、当該撮影画像において特定の物体が写っている領域である対象領域を一次元数値列によって表現したデータである領域記述データと、を対応づけたデータを教師データとして学習させた機械学習モデルを記憶し、撮影画像を上記の機械学習モデルに入力することにより領域記述データを推定する。上記の領域記述データは、撮影画像から背景分離画像を生成し、生成した背景分離画像に基づき生成されたものであり、情報処理システムは、撮影画像を機械学習モデルに入力することにより領域記述データを推定し、領域記述データに基づき背景分離画像を復元する。【選択図】図1
Bibliography:Application Number: JP20190147250