Update README.md by adding a project using maskrcnn-benchmark (#1163)

Thanks for your excellent work! @fmassa 
Based on this repository, we research on the object detectors without sampling heuristics (e.g., Focal Loss, GHM, Undersampling).
The paper (https://arxiv.org/abs/1909.04868) and code (https://github.com/ChenJoya/sampling-free) have been released. Thank you again for maskrcnn-benchmark. It is a really simple, efficient, high performance object detection benchmark.

README.md

@@ -271,6 +271,9 @@ note = {Accessed: [Insert date here]}
 - [MULAN: Multitask Universal Lesion Analysis Network for Joint Lesion Detection, Tagging, and Segmentation](https://arxiv.org/abs/1908.04373).
   Ke Yan, Youbao Tang, Yifan Peng, Veit Sandfort, Mohammadhadi Bagheri, Zhiyong Lu, and Ronald M. Summers.
   MICCAI 2019. [[code](https://github.com/rsummers11/CADLab/tree/master/MULAN_universal_lesion_analysis)]
+- [Is Sampling Heuristics Necessary in Training Deep Object Detectors?](https://arxiv.org/abs/1909.04868)
+  Joya Chen, Dong Liu, Tong Xu, Shilong Zhang, Shiwei Wu, Bin Luo, Xuezheng Peng, Enhong Chen.
+  Tech report, arXiv,1909.04868. [[code](https://github.com/ChenJoya/sampling-free)]
 
 ## License