[paper reading] C-MIL: Continuation Multiple Instance Learning for Weakly Supervised Object Detection CVPR2019

本文轉載自查看原文 2019-04-14 23:52 643

MIL陷入局部最優，檢測到局部，無法完整的檢測到物體。將instance划分為空間相關和類別相關的子集。在這些子集中定義一系列平滑的損失近似代替原損失函數，優化這些平滑損失。

C-MIL learns instance subsets, where the instances are spatially related, i.e., overlapping with each other, and class related, i.e., having similar object class scores.

C-MIL treats images as bags and image regions generated by an object proposal method [24,32] as instances

待解決的問題：

1) How to optimize the non-convex function

2) How to perform instance selection in the early training stages when the instance selector is not well trained.

to be continue ...

更完整的論文筆記[csdn]

免責聲明！

本站轉載的文章為個人學習借鑒使用，本站對版權不負任何法律責任。如果侵犯了您的隱私權益，請聯系本站郵箱yoyou2525@163.com刪除。

猜您在找 [ICCV 2019] Weakly Supervised Object Detection With Segmentation Collaboration 論文閱讀筆記五十三：Libra R-CNN: Towards Balanced Learning for Object Detection(CVPR2019) [CVPR2015] Is object localization for free? – Weakly-supervised learning with convolutional neural networks論文筆記 [CVPR2017] Deep Self-Taught Learning for Weakly Supervised Object Localization 論文筆記多示例學習 multiple instance learning (MIL) Paper Reading: Relation Networks for Object Detection 論文閱讀筆記五十一：CenterNet: Keypoint Triplets for Object Detection(CVPR2019) 論文閱讀筆記四十八：Bounding Box Regression with Uncertainty for Accurate Object Detection(CVPR2019) 論文閱讀筆記五十七:FCOS: Fully Convolutional One-Stage Object Detection(CVPR2019) 論文閱讀筆記五十二：CornerNet-Lite: Efficient Keypoint Based Object Detection（CVPR2019）