전체 글92 [논문리뷰] Zero-Shot Learning on 3D Point Cloud Objects and Beyond Paper OverviewIJCV'22https://arxiv.org/abs/2104.04980 Zero-Shot Learning on 3D Point Cloud Objects and BeyondZero-shot learning, the task of learning to recognize new classes not seen during training, has received considerable attention in the case of 2D image classification. However, despite the increasing ubiquity of 3D sensors, the corresponding 3D point cloudarxiv.org Abstract저자들은 3D ZSL에 대한.. 2024. 4. 16. [논문리뷰] FREE: Feature Refinement for Generalized Zero-Shot Learning Paper Overview ICCV'21 https://arxiv.org/abs/2107.13807 FREE: Feature Refinement for Generalized Zero-Shot Learning Generalized zero-shot learning (GZSL) has achieved significant progress, with many efforts dedicated to overcoming the problems of visual-semantic domain gap and seen-unseen bias. However, most existing methods directly use feature extraction models traine arxiv.org Abstract 대부분 존재.. 2024. 4. 11. [논문리뷰] Adaptive Confidence Smoothing for Generalized Zero-Shot Learning Paper Overview CVPR'19 https://arxiv.org/abs/1812.09903 Adaptive Confidence Smoothing for Generalized Zero-Shot Learning Generalized zero-shot learning (GZSL) is the problem of learning a classifier where some classes have samples and others are learned from side information, like semantic attributes or text description, in a zero-shot learning fashion (ZSL). Training a sing arxiv.org Abstract 저.. 2024. 3. 25. [논문리뷰] A Semantic Encoding Out-of-Distribution Classifier for Generalized Zero-Shot Learning Paper Overview IEEE Signal Processing Letters'21 https://ieeexplore.ieee.org/document/9464653 Request Rejected ieeexplore.ieee.org Abstract 저자들은 새로운 semantic encoding out-of-distribution classifier (SE-OOD)를 제안한다. 본 방법은 먼저 semantically consistent mapping을 하여 모든 visual sample을 대응되는 semantic attribute에 project한다. 그 다음 projected visual sample과 원래 semantic attribute 둘다 distribution alignment를 위해 lat.. 2024. 3. 18. Pytorch Learnable parameter, 학습가능한 파라미터 만들기 우리가 pytorch 내에서 학습 가능한 파라미터를 만들어야 할 때가 종종 있다. 1 이때 nn.Module 내에 파라미터를 선언할 경우 parameter = nn.Parameter(torch.randn(10).cuda()) 위와 같이 선언하여 사용하면 된다. 2 nn.Module 외에서 위 파라미터를 선언한 후 아래와 같이 연산을 한다면 parameter = parameter.exp() parameter = parameter.double() 아래와 같은 ValueError: can't optimize a non-leaf Tensor 에러를 만나게 된다. 이는 이미 만들어진 파라미터를 다른연산으로 casting하였기 때문이라고 하는데 pytorch를 자세히 공부하지 않아서 이해는 안된다. 하지만 해결 방.. 2024. 3. 14. [논문리뷰] Transductive Zero-Shot Learning for 3D Point Cloud Classification Paper OverviewWACV'20https://arxiv.org/abs/1912.07161 Transductive Zero-Shot Learning for 3D Point Cloud ClassificationZero-shot learning, the task of learning to recognize new classes not seen during training, has received considerable attention in the case of 2D image classification. However despite the increasing ubiquity of 3D sensors, the corresponding 3D point cloudarxiv.org Abstract이 논문은 .. 2024. 3. 12. 이전 1 2 3 4 5 6 7 8 ··· 16 다음
