# An Implementation of Faster RCNN with Study for Region Sampling

arXiv: https://arxiv.org/abs/1702.02138 · HTML: https://mlsanity.com/paper/1702.02138/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Feb 2017.

## Built on by (most cited first, 7 of 7)

- [ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation](https://mlsanity.com/paper/2204.12484/) (2022)
- [UTD-Yolov5: A Real-time Underwater Targets Detection Method based on Attention Improved YOLOv5](https://mlsanity.com/paper/2207.00837/) (2022)
- [Advanced Efficient Strategy for Detection of Dark Objects Based on Spiking Network with Multi-Box Detection](https://mlsanity.com/paper/2310.06370/) (2023)
- [Deep Convolutional Neural Network for Plume Rise Measurements in Industrial Environments](https://mlsanity.com/paper/2302.07416/) (2023)
- [Implementing a Real-Time, YOLOv5 based Social Distancing Measuring System for Covid-19](https://mlsanity.com/paper/2204.03350/) (2022)
- [Improving Performance of Object Detection using the Mechanisms of Visual Recognition in Humans](https://mlsanity.com/paper/2301.09667/) (2023)
- [Refinement Module based on Parse Graph for Human Pose Estimation](https://mlsanity.com/paper/2501.11069/) (2025)

## Abstract (arXiv)

We adapted the join-training scheme of Faster RCNN framework from Caffe to TensorFlow as a baseline implementation for object detection. Our code is made publicly available. This report documents the simplifications made to the original pipeline, with justifications from ablation analysis on both PASCAL VOC 2007 and COCO 2014. We further investigated the role of non-maximal suppression (NMS) in selecting regions-of-interest (RoIs) for region classification, and found that a biased sampling toward small regions helps performance and can achieve on-par mAP to NMS-based sampling when converged sufficiently.