TestK

논문 제목 : Stitch it in Time: GAN-Based Facial Editing of Real Videos 논문 주소 : https://arxiv.org/abs/2201.08361 Youtube video : https://www.youtube.com/watch?v=4lQkQSmA8nA Stitch it in Time: GAN-Based Facial Editing of Real Videos The ability of Generative Adversarial Networks to encode rich semantics within their latent space has been widely adopted for facial image editing. However, replicating th..

논문 제목: CAMERAS: Enhanced Resolution And Sanity Preserving Class Activation Mapping For Image Saliency 논문 주소: https://arxiv.org/abs/2106.10649 CAMERAS: Enhanced Resolution And Sanity preserving Class Activation Mapping for image saliency Backpropagation image saliency aims at explaining model predictions by estimating model-centric importance of individual pixels in the input. However, class-inse..

논문 제목 : Towards Learning Spatially Discriminative Feature Representation 논문 주소 : https://arxiv.org/abs/2109.01359 Towards Learning Spatially Discriminative Feature Representations The backbone of traditional CNN classifier is generally considered as a feature extractor, followed by a linear layer which performs the classification. We propose a novel loss function, termed as CAM-loss, to constrai..

논문 제목 : Informative Class Activation Maps 논문 주소 : https://arxiv.org/abs/2106.10472 Informative Class Activation Maps We study how to evaluate the quantitative information content of a region within an image for a particular label. To this end, we bridge class activation maps with information theory. We develop an informative class activation map (infoCAM). Given a classi arxiv.org 주요 내용 정리: 1) 저..

논문 제목 : Ablation-CAM: Visual Explanations for Deep Convolutional Network Via Gradient-free Localization 논문 주소 : https://openaccess.thecvf.com/content_WACV_2020/html/Desai_Ablation-CAM_Visual_Explanations_for_Deep_Convolutional_Network_via_Gradient-free_Localization_WACV_2020_paper.html WACV 2020 Open Access Repository Ablation-CAM: Visual Explanations for Deep Convolutional Network via Gradient-..

논문 제목 : SmoothGrad : removing noise by adding noise 논문 주소 : arxiv.org/abs/1706.03825 SmoothGrad: removing noise by adding noise Explaining the output of a deep network remains a challenge. In the case of an image classifier, one type of explanation is to identify pixels that strongly influence the final decision. A starting point for this strategy is the gradient of the class score arxiv.org 주요 ..

논문 제목 : Smooth Grad-CAM++: An Enhanced Inference Level Visualization Technique for Deep Convolutional Neural Network Models 논문 주소 : arxiv.org/abs/1908.01224 Smooth Grad-CAM++: An Enhanced Inference Level Visualization Technique for Deep Convolutional Neural Network Models Gaining insight into how deep convolutional neural network models perform image classification and how to explain their outpu..

논문 제목 : Grad-CAM++: Generalized Gradient-based Visual Explanations for Deep Convolutional Networks 논문 주소 : arxiv.org/pdf/1710.11063.pdf IEEE WACV (2018, ieeexplore.ieee.org/document/8354201)에 나온 논문을 바탕으로 이해하고 내용을 작성한다. arixv에서 나온 버전이 좀 더 extended version임으로 Grad-CAM++에 더 깊은 이해를 위해서는 extended version을 읽는 것을 추천한다. 주요 내용 : 1) Deep models은 "black box"로서 internal function을 이해하는데에는 어려움이 있다. 이를 해결하기 위해..

논문 제목 : Learning deep features for discriminative localization 논문 주소 : openaccess.thecvf.com/content_iccv_2017/html/Selvaraju_Grad-CAM_Visual_Explanations_ICCV_2017_paper.html ICCV 2017 Open Access Repository Grad-CAM: Visual Explanations From Deep Networks via Gradient-Based Localization Ramprasaath R. Selvaraju, Michael Cogswell, Abhishek Das, Ramakrishna Vedantam, Devi Parikh, Dhruv Batra; Pr..

논문 제목 : Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization 논문 주소 : openaccess.thecvf.com/content_iccv_2017/html/Selvaraju_Grad-CAM_Visual_Explanations_ICCV_2017_paper.html ICCV 2017 Open Access Repository Grad-CAM: Visual Explanations From Deep Networks via Gradient-Based Localization Ramprasaath R. Selvaraju, Michael Cogswell, Abhishek Das, Ramakrishna Vedantam, De..