# Equalization Loss v2: A New Gradient Balance Approach for Long-tailed Object Detection

arXiv: https://arxiv.org/abs/2012.08548 · HTML: https://mlsanity.com/paper/2012.08548/

## Key facts

- Cited by 99 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 15 Dec 2020.

## Built on by (most cited first, 30 of 99)

- [Objaverse: A Universe of Annotated 3D Objects](https://mlsanity.com/paper/2212.08051/) (2022)
- [Detecting Twenty-thousand Classes using Image-level Supervision](https://mlsanity.com/paper/2201.02605/) (2022)
- [End-to-end Autonomous Driving: Challenges and Frontiers](https://mlsanity.com/paper/2306.16927/) (2023)
- [Deep Long-Tailed Learning: A Survey](https://mlsanity.com/paper/2110.04596/) (2021)
- [Towards Open Vocabulary Learning: A Survey](https://mlsanity.com/paper/2306.15880/) (2023)
- [ACE: Ally Complementary Experts for Solving Long-Tailed Recognition in One-Shot](https://mlsanity.com/paper/2108.02385/) (2021)
- [FASA: Feature Augmentation and Sampling Adaptation for Long-Tailed Instance Segmentation](https://mlsanity.com/paper/2102.12867/) (2021)
- [A Survey on Long-Tailed Visual Recognition](https://mlsanity.com/paper/2205.13775/) (2022)
- [Detecting Everything in the Open World: Towards Universal Object Detection](https://mlsanity.com/paper/2303.11749/) (2023)
- [Exploring Classification Equilibrium in Long-Tailed Object Detection](https://mlsanity.com/paper/2108.07507/) (2021)
- [DiffusionInst: Diffusion Model for Instance Segmentation](https://mlsanity.com/paper/2212.02773/) (2022)
- [X-Paste: Revisiting Scalable Copy-Paste for Instance Segmentation using CLIP and StableDiffusion](https://mlsanity.com/paper/2212.03863/) (2022)
- [Equalized Focal Loss for Dense Long-Tailed Object Detection](https://mlsanity.com/paper/2201.02593/) (2022)
- [On Model Calibration for Long-Tailed Object Detection and Instance Segmentation](https://mlsanity.com/paper/2107.02170/) (2021)
- [Disentangle Your Dense Object Detector](https://mlsanity.com/paper/2107.02963/) (2021)
- [Gradient Reweighting: Towards Imbalanced Class-Incremental Learning](https://mlsanity.com/paper/2402.18528/) (2024)
- [MosaicFusion: Diffusion Models as Data Augmenters for Large Vocabulary Instance Segmentation](https://mlsanity.com/paper/2309.13042/) (2023)
- [Invariant Feature Learning for Generalized Long-Tailed Classification](https://mlsanity.com/paper/2207.09504/) (2022)
- [MosaicOS: A Simple and Effective Use of Object-Centric Images for Long-Tailed Object Detection](https://mlsanity.com/paper/2102.08884/) (2021)
- [A Systematic Review on Long-Tailed Learning](https://mlsanity.com/paper/2408.00483/) (2024)
- [Open Vocabulary Object Detection with Proposal Mining and Prediction Equalization](https://mlsanity.com/paper/2206.11134/) (2022)
- [Rank & Sort Loss for Object Detection and Instance Segmentation](https://mlsanity.com/paper/2107.11669/) (2021)
- [Use Your Head: Improving Long-Tail Video Recognition](https://mlsanity.com/paper/2304.01143/) (2023)
- [Improving the Intra-class Long-tail in 3D Detection via Rare Example Mining](https://mlsanity.com/paper/2210.08375/) (2022)
- [SFC: Shared Feature Calibration in Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/2401.11719/) (2024)
- [Relieving Long-tailed Instance Segmentation via Pairwise Class Balance](https://mlsanity.com/paper/2201.02784/) (2022)
- [Towards Calibrated Hyper-Sphere Representation via Distribution Overlap Coefficient for Long-tailed Learning](https://mlsanity.com/paper/2208.10043/) (2022)
- [Fed-GraB: Federated Long-tailed Learning with Self-Adjusting Gradient Balancer](https://mlsanity.com/paper/2310.07587/) (2023)
- [Label-Aware Distribution Calibration for Long-tailed Classification](https://mlsanity.com/paper/2111.04901/) (2021)
- [Long-tail Detection with Effective Class-Margins](https://mlsanity.com/paper/2301.09724/) (2023)

## Abstract (arXiv)

Recently proposed decoupled training methods emerge as a dominant paradigm for long-tailed object detection. But they require an extra fine-tuning stage, and the disjointed optimization of representation and classifier might lead to suboptimal results. However, end-to-end training methods, like equalization loss (EQL), still perform worse than decoupled training methods. In this paper, we reveal the main issue in long-tailed object detection is the imbalanced gradients between positives and negatives, and find that EQL does not solve it well. To address the problem of imbalanced gradients, we introduce a new version of equalization loss, called equalization loss v2 (EQL v2), a novel gradient guided reweighing mechanism that re-balances the training process for each category independently and equally. Extensive experiments are performed on the challenging LVIS benchmark. EQL v2 outperforms origin EQL by about 4 points overall AP with 14-18 points improvements on the rare categories. More importantly, it also surpasses decoupled training methods. Without further tuning for the Open Images dataset, EQL v2 improves EQL by 7.3 points AP, showing strong generalization ability. Codes have been released at https://github.com/tztztztztz/eqlv2