폐색 이미지 분류를 위한 강건한 가중치 전환 학습
The Robust Weight Conversion Learning for Classification of Occlusion Images
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
An unexpected occlusion in a real life, not in a laboratory, can be more fatal to neural networks than expected. In addition, it is virtually impossible to create a network that learns all the environmental changes as well as occlusions. Therefore, we propose an alternative approach in which the architecture and number of parameters remain unchanged while adapting to occlusion circumstances. Learning method with the term Conversion Learning classifies them more robustly by converting the weights from various occlusion situations. The experiments on MNIST dataset showed a 3.07 [%p] performance improvement over the baseline CNN model in a situation where most objects are occluded and unknowing what occlusion will appear in advance. The experimental results suggest that Conversion Learning is an efficient method to respond to environmental changes such as occluded images.
Abstract
An unexpected occlusion in a real life, not in a laboratory, can be more fatal to neural networks than expected. In addition, it is virtually impossible to create a network that learns all the environmental changes as well as occlusions. Therefore, we propose an alternative approach in which the architecture and number of parameters remain unchanged while adapting to occlusion circumstances. Learning method with the term Conversion Learning classifies them more robustly by converting the weights from various occlusion situations. The experiments on MNIST dataset showed a 3.07 [%p] performance improvement over the baseline CNN model in a situation where most objects are occluded and unknowing what occlusion will appear in advance. The experimental results suggest that Conversion Learning is an efficient method to respond to environmental changes such as occluded images.
참고문헌 (14)
- S. O’Gara, "Comparing data augmentation strategies for deep image classification", IMVIP 2019: Irish Machine Vision and Image Processing (IMVIP) (2019)
- A. Bochkovskiy, "Yolov4 : Op-timal speed and accuracy of object detection", Computer Vision and Pattern Recognition (2020)
- Z. Zhong, "Random erasing data augmentation", AAAI Conference on Artificial Intelligence, 34(7) (2020)
- T. DeVries, "Improved regularization of convolutional neural networks with cutout", Computer Vision and Pattern Recognition (2017)
- Krishna Kumar Singh, "Hide-and-Seek: Forcing a Network to be Meticulous for Weakly-Supervised Object and Action Localization", 2017 IEEE International Conference on Computer Vision (ICCV) (2017), IEEE
- P. Chen, "Gridmask data augmentation", Computer Vision and Pattern Recognition (2020)
- D. Walawalkar, "Attentive cutmix : An enhanced data augmentation approach for deep learning based image classification", Computer Vision and Pattern Recognition (2020)
- Dan Zeng, "A survey of face recognition techniques under occlusion", IET Biometrics, 10(6), pp.581~606 (2021), Institution of Engineering and Technology (IET)[DOI]
- Lingxue Song, "Occlusion Robust Face Recognition Based on Mask Learning With Pairwise Differential Siamese Network", 2019 IEEE, 2019, ~[DOI]
- Feng Cen, "Deep feature augmentation for occluded image classification", Pattern Recognition, 111, pp.107737~ (2021), Elsevier BV[DOI]
- "The mnist database of handwritten digits" (1998)[URL]
- "Learning multiple layers of features from tiny images" (2009)[URL]
- D. P. Kingma, "Adam : A method for stochastic optimization", Machine Learning (2014)
- T. Tieleman, "Lecture 6. 5-rmsprop : Divide the gradient by a running average of its recent magnitude", COURSERA: Neural networks for machine learning, 4(2), pp.26~31 (2012)