Awesome
A Repository of the Papers Addressing Imbalance Problems in Object Detection
This repository provides an up-to-date the list of studies addressing imbalance problems in object detection. It follows the taxonomy provided in the following paper (please cite the paper if you benefit from this repository):
K. Oksuz, B. C. Cam, S. Kalkan, E. Akbas, "Imbalance Problems in Object Detection: A Review", Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2020.[preprint]
BibTeX entry:
@ARTICLE{imbalance,
author = {Kemal Oksuz and Baris Can Cam and Sinan Kalkan and Emre Akbas},
title = "{Imbalance Problems in Object Detection: A Review}",
journal = {Transactions on Pattern Analysis and Machine Intelligence (TPAMI)},
year = "2020",
pages = {1-1}
}
How to request addition of a paper
If you know of a paper that addresses an imbalance problem concerning generic object detection and is not on this repository, you are welcome to request the addition of that paper by submitting a pull request. In your pull request please briefly state which section of your paper is related to which problem.
Following the methodology in our paper, the papers should be designed for the generic object detection problem (i.e. reporting results on generic object detection datasets such as ILSVRC, Pascal VOC, MS-COCO, Open Images, Objects 365 etc.).
Table of Contents (Follows the taxonomy in the paper)
- Class Imbalance
1.1 Foreground-Background Class Imbalance
1.2 Foreground-Foreground Class Imbalance - Scale Imbalance
2.1 Object/box-level Scale Imbalance
2.2 Feature-level Imbalance - Spatial Imbalance
3.1 Imbalance in Regression Loss
3.2 IoU Distribution Imbalance
3.3 Object Location Imbalance - Objective Imbalance
1. Class Imbalance <a name="1"></a>
1.1. Foreground-Background Class Imbalance <a name="1.1"></a>
-
Hard Sampling Methods
- Random Sampling
- Hard Example Mining
- Limit Search Space
-
Soft Sampling Methods
-
Sampling-Free Methods
-
Generative Methods
1.2. Foreground-Foreground Class Imbalance <a name="1.2"></a>
- Fine-tuning Long Tail Distribution for Obj.Det., CVPR 2016, [paper]
- PSIS, arXiv 2019, [paper]
- OFB Sampling, WACV 2020, [paper]
- Large-Scale Object Detection in the Wild from Imbalanced Multi-Labels, CVPR 2020, [paper]
- Balanced Group Softmax Loss, CVPR 2020, [paper]
- SimCal, ECCV 2020, [paper]
2. Scale Imbalance <a name="2"></a>
2.1. Object/box-level Scale Imbalance <a name="2.1"></a>
-
Methods Predicting from the Feature Hierarchy of Backbone Features
-
Methods Based on Feature Pyramids
- FPN, CVPR 2017, [paper]
- See feature-level imbalance methods
-
Methods Based on Image Pyramids
-
Methods Combining Image and Feature Pyramids
2.2. Feature-level Imbalance <a name="2.2"></a>
-
Methods Using Pyramidal Features as a Basis
-
Methods Using Backbone Features as a Basis
- STDN, CVPR 2018, [paper]
- Parallel-FPN, ECCV 2018, [paper]
- Deep Feature Pyramid Reconfiguration, ECCV 2018, [paper]
- Zoom Out-and-In, IJCV 2019, [paper]
- Multi-level FPN, AAAI 2019, [paper]
- NAS-FPN, CVPR 2019, [paper]
- Auto-FPN, ICCV 2019, [paper]
- AugFPN, CVPR 2020, [paper]
- Hit-Detector, CVPR 2020, [paper]
- Pyramid Convolution, CVPR 2020, [paper]
3. Spatial Imbalance <a name="3"></a>
3.1. Imbalance in Regression Loss <a name="3.1"></a>
-
Lp norm based
-
IoU based
-
Cross entropy based
3.2. IoU Distribution Imbalance <a name="3.2"></a>
- Cascade R-CNN, CVPR 2018, [paper]
- Hierarchical Shot Detector, ICCV 2019, [paper]
- IoU-uniform R-CNN, arXiv 2019, [paper]
- pRoI Generator, WACV 2020, [paper]
3.3. Object Location Imbalance <a name="3.3"></a>
4. Objective Imbalance <a name="4"></a>
- Task Weighting
- Classification Aware Regression Loss, CVPR 2020, [paper]
- Guided Loss, arXiv 2019, [paper]
- aLRP Loss, NeurIPS 2020, [paper]
- RS Loss, ICCV 2021, [paper]
Contact
Please contact Kemal Öksüz (kemal.oksuz@metu.edu.tr) for your questions about this webpage.