# An Implementation of Faster RCNN with Study for Region Sampling

Source: [An Implementation of Faster RCNN with Study for Region Sampling](<https://www.sota2.com/research/paper/68fa63201e209131ce9fd112>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T20:34:16.778Z

Authors: Xinlei Chen, Abhinav Gupta

Published: 2017-02-07T18:54:34.000Z

## Abstract

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.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Object Detection on PASCAL VOC 2007 (test) including Efficiency Metrics](<https://www.sota2.com/research/sota/object-detection-on-pascal-voc-2007-test-including-efficiency-metrics.md>): Object Detection; PASCAL VOC 2007 (test); mAP: 79.8.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/1702.02138>)

[PDF](<https://arxiv.org/pdf/1702.02138.pdf>)