NIPS 2019 Meta-Reinforced Synthetic Data for One-Shot Fine-Grained Visual Recognition
文章目录 Motivation Method Experiment 问题 参考 Motivation 细粒度识别问题 比普通的分类更难 很难收集数据：需要专家标注 需要小样本学习 **是否可以生成更多的数据？**现有的最好的GANS如何？挑战 GAN训练需要许多的数据。可否降低训练的数据量？或者换一种训练方式？ Idea：微调 ImageNet上训练的GANS。 Method Fine-tune
相关文章
- 1. CNN for Visual Recognition (assignment1_Q1)
- 2. CNN for Visual Recognition (02)
- 3. CNN for Visual Recognition (01)
- 4. Learning from Synthetic Data: Addressing Domain Shift for Semantic Segmentation
- 5. CS231n Convolutional Neural Networks for Visual Recognition
- 6. 数据合成方法 ： 论文 Synthetic Data and Artificial Neural Networks for Natural Scene Text Recognition
- 7. nips 208 visual question answering 导读
- 8. NIPS 2019 Transductive Zero-Shot Learning with Visual Structure Constraint
- 9. Convolutional Neural Networks for Visual Recognition
- 10. [Synthetic-data-with-text-and-image]
- 更多相关文章...
- • Swift for 循环 - Swift 教程
- • Scala for循环 - Scala教程
- • JDK13 GA发布：5大特性解读
- • Java Agent入门实战（三）-JVM Attach原理与使用
相关标签/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
欢迎关注本站公众号,获取更多信息
相关文章
- 1. CNN for Visual Recognition (assignment1_Q1)
- 2. CNN for Visual Recognition (02)
- 3. CNN for Visual Recognition (01)
- 4. Learning from Synthetic Data: Addressing Domain Shift for Semantic Segmentation
- 5. CS231n Convolutional Neural Networks for Visual Recognition
- 6. 数据合成方法 ： 论文 Synthetic Data and Artificial Neural Networks for Natural Scene Text Recognition
- 7. nips 208 visual question answering 导读
- 8. NIPS 2019 Transductive Zero-Shot Learning with Visual Structure Constraint
- 9. Convolutional Neural Networks for Visual Recognition
- 10. [Synthetic-data-with-text-and-image]