# Recursively Refined R-CNN: Instance Segmentation with Self-RoI Rebalancing

arXiv: https://arxiv.org/abs/2104.01329 · HTML: https://mlsanity.com/paper/2104.01329/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 27 references that are in the map.
- First posted on arXiv 3 Apr 2021.

## Builds on (most cited first, 23 of 23)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Mask R-CNN](https://mlsanity.com/paper/1703.06870/) (2017)
- [You Only Look Once: Unified, Real-Time Object Detection](https://mlsanity.com/paper/1506.02640/) (2015)
- [Non-local Neural Networks](https://mlsanity.com/paper/1711.07971/) (2017)
- [Rich feature hierarchies for accurate object detection and semantic segmentation](https://mlsanity.com/paper/1311.2524/) (2013)
- [MMDetection: Open MMLab Detection Toolbox and Benchmark](https://mlsanity.com/paper/1906.07155/) (2019)
- [Cascade R-CNN: Delving into High Quality Object Detection](https://mlsanity.com/paper/1712.00726/) (2017)
- [Deformable ConvNets v2: More Deformable, Better Results](https://mlsanity.com/paper/1811.11168/) (2018)
- [Training Region-based Object Detectors with Online Hard Example Mining](https://mlsanity.com/paper/1604.03540/) (2016)
- [Hybrid Task Cascade for Instance Segmentation](https://mlsanity.com/paper/1901.07518/) (2019)
- [GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond](https://mlsanity.com/paper/1904.11492/) (2019)
- [Mask Scoring R-CNN](https://mlsanity.com/paper/1903.00241/) (2019)
- [Libra R-CNN: Towards Balanced Learning for Object Detection](https://mlsanity.com/paper/1904.02701/) (2019)
- [Imbalance Problems in Object Detection: A Review](https://mlsanity.com/paper/1909.00169/) (2019)
- [Region Proposal by Guided Anchoring](https://mlsanity.com/paper/1901.03278/) (2019)
- [Cascade RPN: Delving into High-Quality Region Proposal Network with Adaptive Convolution](https://mlsanity.com/paper/1909.06720/) (2019)
- [Revisiting RCNN: On Awakening the Classification Power of Faster RCNN](https://mlsanity.com/paper/1803.06799/) (2018)
- [A novel Region of Interest Extraction Layer for Instance Segmentation](https://mlsanity.com/paper/2004.13665/) (2020)
- [Matching-CNN Meets KNN: Quasi-Parametric Human Parsing](https://mlsanity.com/paper/1504.01220/) (2015)
- [IoU-uniform R-CNN: Breaking Through the Limitations of RPN](https://mlsanity.com/paper/1912.05190/) (2019)
- [Generating Positive Bounding Boxes for Balanced Training of Object Detectors](https://mlsanity.com/paper/1909.09777/) (2019)
- [Cascade Region Proposal and Global Context for Deep Object Detection](https://mlsanity.com/paper/1710.10749/) (2017)

## Built on by (most cited first, 5 of 5)

- [Empowering DINO Representations for Underwater Instance Segmentation via Aligner and Prompter](https://mlsanity.com/paper/2511.08334/) (2025)
- [UIS-Mamba: Exploring Mamba for Underwater Instance Segmentation via Dynamic Tree Scan and Hidden State Weaken](https://mlsanity.com/paper/2508.00421/) (2025)
- [LDD: A Dataset for Grape Diseases Object Detection and Instance Segmentation](https://mlsanity.com/paper/2206.10192/) (2022)
- [Self-Balanced R-CNN for Instance Segmentation](https://mlsanity.com/paper/2404.16633/) (2024)
- [ZMIS-SAM: Segment Anything Model Enhanced with Wavelet Transform for Zooplankton Microscopy Image Instance Segmentation](https://mlsanity.com/paper/2607.27585/) (2026)

## Abstract (arXiv)

Within the field of instance segmentation, most of the state-of-the-art deep learning networks rely nowadays on cascade architectures, where multiple object detectors are trained sequentially, re-sampling the ground truth at each step. This offers a solution to the problem of exponentially vanishing positive samples. However, it also translates into an increase in network complexity in terms of the number of parameters. To address this issue, we propose Recursively Refined R-CNN (R^3-CNN) which avoids duplicates by introducing a loop mechanism instead. At the same time, it achieves a quality boost using a recursive re-sampling technique, where a specific IoU quality is utilized in each recursion to eventually equally cover the positive spectrum. Our experiments highlight the specific encoding of the loop mechanism in the weights, requiring its usage at inference time. The R^3-CNN architecture is able to surpass the recently proposed HTC model, while reducing the number of parameters significantly. Experiments on COCO minival 2017 dataset show performance boost independently from the utilized baseline model. The code is available online at https://github.com/IMPLabUniPr/mmdetection/tree/r3_cnn.