论文地址：UberNet: Training a ‘Universal’ Convolutional Neural Network for Low-, Mid-, and High-Level Vision using Diverse Datasets and Limited Memory 博客参考：https://blog.csdn.net/u012905422/article/details/52507100 项目参考：https://github.com/jkokkin/UberNet/tree/master/caffe-fast-rcnn
解决目标
- 将不同任务通过同一个网络进行联合训练
- 将1中的目的在尽量少的内存中进行实现
Through these advances we train in an end-to-end manner a CNN that simultaneously addresses (a) _boundary detection (b) normal estimation (c) saliency estimation (d) semantic segmentation (e) human part segmentation (f) semantic boundary detection, (g)_ region proposal generation and object detection.
网络架构
结果
若有收获，就点个赞吧
0 人点赞
