CAM-BASED WEAKLY SUPERVISED LEARNING OBJECT LOCALIZATION DEVICE AND METHOD
To provide CAM-based weakly supervised learning object localization device and method that improve object localization performance of weakly supervised learning.SOLUTION: The present invention relates to CAM-based weakly supervised object localization device and method. The device includes: a featur...
Saved in:
Main Authors | , , , , |
---|---|
Format | Patent |
Language | English Japanese |
Published |
04.04.2023
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | To provide CAM-based weakly supervised learning object localization device and method that improve object localization performance of weakly supervised learning.SOLUTION: The present invention relates to CAM-based weakly supervised object localization device and method. The device includes: a feature map extractor for extracting a feature map of a last convolutional layer in a CNN in a process of applying an image to the CNN; a weight vector binarization unit for first binarizing a weight vector of a linear layer in a process of sequentially applying the feature map to a pooling layer that generates a feature vector and the linear layer that generates a class label; a feature map binarization unit for second binarizing the feature map on the basis of the first binarized weight vector; and a class activation map generation unit for generating a class activation map for object localization on the basis of the second binarized feature map.SELECTED DRAWING: Figure 2
【課題】弱教師あり学習の物体探知性能を向上させるCAM基盤の弱教師あり学習物体探知装置及び方法を提供する。【解決手段】本発明は、CAM基盤の弱教師あり学習物体探知装置及び方法に関し、前記装置は、イメージをCNNに適用する過程でCNNにある最後のコンボリューションレイヤのフィーチャマップを抽出するフィーチャマップ抽出部と、フィーチャマップを、フィーチャベクトルを生成するプーリングレイヤとクラスラベルを生成するリニアレイヤとに順次適用する過程でリニアレイヤの加重値ベクトルを第1二値化する加重値ベクトル二値化部と、第1二値化された加重値ベクトルを基にフィーチャマップを第2二値化するフィーチャマップ二値化部と、第2二値化されたフィーチャマップを基に物体探知のためのクラス活性化マップを生成するクラス活性化マップ生成部と、を備える。【選択図】図2 |
---|---|
Bibliography: | Application Number: JP20210196551 |