# Beyond Max-Margin: Class Margin Equilibrium for Few-shot Object Detection

arXiv: https://arxiv.org/abs/2103.04612 · HTML: https://mlsanity.com/paper/2103.04612/

## Key facts

- Cited by 80 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 37 references that are in the map.
- First posted on arXiv 8 Mar 2021.

## Builds on (most cited first, 26 of 26)

- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [Feature Pyramid Networks for Object Detection](https://mlsanity.com/paper/1612.03144/) (2016)
- [Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks](https://mlsanity.com/paper/1703.03400/) (2017)
- [You Only Look Once: Unified, Real-Time Object Detection](https://mlsanity.com/paper/1506.02640/) (2015)
- [SSD: Single Shot MultiBox Detector](https://mlsanity.com/paper/1512.02325/) (2015)
- [Fast R-CNN](https://mlsanity.com/paper/1504.08083/) (2015)
- [Matching Networks for One Shot Learning](https://mlsanity.com/paper/1606.04080/) (2016)
- [YOLO9000: Better, Faster, Stronger](https://mlsanity.com/paper/1612.08242/) (2016)
- [Learning to Compare: Relation Network for Few-Shot Learning](https://mlsanity.com/paper/1711.06025/) (2017)
- [Frustratingly Simple Few-Shot Object Detection](https://mlsanity.com/paper/2003.06957/) (2020)
- [Few-shot Object Detection via Feature Reweighting](https://mlsanity.com/paper/1812.01866/) (2018)
- [Prototype Mixture Models for Few-shot Semantic Segmentation](https://mlsanity.com/paper/2008.03898/) (2020)
- [Meta R-CNN : Towards General Solver for Instance-level Few-shot Learning](https://mlsanity.com/paper/1909.13032/) (2019)
- [Low-Shot Learning from Imaginary Data](https://mlsanity.com/paper/1801.05401/) (2018)
- [Task-Agnostic Meta-Learning for Few-shot Learning](https://mlsanity.com/paper/1805.07722/) (2018)
- [Few-Shot Object Detection and Viewpoint Estimation for Objects in the Wild](https://mlsanity.com/paper/2007.12107/) (2020)
- [Negative Margin Matters: Understanding Margin in Few-shot Classification](https://mlsanity.com/paper/2003.12060/) (2020)
- [Multi-Scale Positive Sample Refinement for Few-Shot Object Detection](https://mlsanity.com/paper/2007.09384/) (2020)
- [Adversarial Feature Hallucination Networks for Few-Shot Learning](https://mlsanity.com/paper/2003.13193/) (2020)
- [Boosting Few-Shot Learning With Adaptive Margin Loss](https://mlsanity.com/paper/2005.13826/) (2020)
- [LSTD: A Low-Shot Transfer Detector for Object Detection](https://mlsanity.com/paper/1803.01529/) (2018)
- [C-MIL: Continuation Multiple Instance Learning for Weakly Supervised Object Detection](https://mlsanity.com/paper/1904.05647/) (2019)
- [Learning Compositional Representations for Few-Shot Recognition](https://mlsanity.com/paper/1812.09213/) (2018)
- [Min-Entropy Latent Model for Weakly Supervised Object Detection](https://mlsanity.com/paper/1902.06057/) (2019)
- [Soft Proposal Networks for Weakly Supervised Object Localization](https://mlsanity.com/paper/1709.01829/) (2017)
- [Learning to Select Base Classes for Few-shot Classification](https://mlsanity.com/paper/2004.00315/) (2020)

## Built on by (most cited first, 30 of 80)

- [Prompt, Generate, then Cache: Cascade of Foundation Models makes Strong Few-shot Learners](https://mlsanity.com/paper/2303.02151/) (2023)
- [Hallucination Improves Few-Shot Object Detection](https://mlsanity.com/paper/2105.01294/) (2021)
- [Meta-DETR: Image-Level Few-Shot Detection with Inter-Class Correlation Exploitation](https://mlsanity.com/paper/2208.00219/) (2022)
- [Meta-DETR: Image-Level Few-Shot Object Detection with Inter-Class Correlation Exploitation](https://mlsanity.com/paper/2103.11731/) (2021)
- [Label, Verify, Correct: A Simple Few Shot Object Detection Method](https://mlsanity.com/paper/2112.05749/) (2021)
- [Few-Shot Object Detection via Association and DIscrimination](https://mlsanity.com/paper/2111.11656/) (2021)
- [Explore the Power of Synthetic Data on Few-shot Object Detection](https://mlsanity.com/paper/2303.13221/) (2023)
- [A Survey of Self-Supervised and Few-Shot Object Detection](https://mlsanity.com/paper/2110.14711/) (2021)
- [Revisiting Few-Shot Object Detection with Vision-Language Models](https://mlsanity.com/paper/2312.14494/) (2023)
- [Few-Shot Object Detection: A Comprehensive Survey](https://mlsanity.com/paper/2112.11699/) (2021)
- [Anti-aliasing Semantic Reconstruction for Few-Shot Semantic Segmentation](https://mlsanity.com/paper/2106.00184/) (2021)
- [Sylph: A Hypernetwork Framework for Incremental Few-shot Object Detection](https://mlsanity.com/paper/2203.13903/) (2022)
- [Few-Shot Object Detection via Variational Feature Aggregation](https://mlsanity.com/paper/2301.13411/) (2023)
- [Task Discrepancy Maximization for Fine-grained Few-Shot Classification](https://mlsanity.com/paper/2207.01376/) (2022)
- [DiGeo: Discriminative Geometry-Aware Learning for Generalized Few-Shot Object Detection](https://mlsanity.com/paper/2303.09674/) (2023)
- [Generating Features with Increased Crop-related Diversity for Few-Shot Object Detection](https://mlsanity.com/paper/2304.05096/) (2023)
- [FS-DETR: Few-Shot DEtection TRansformer with prompting and without re-training](https://mlsanity.com/paper/2210.04845/) (2022)
- [Semantic-aligned Fusion Transformer for One-shot Object Detection](https://mlsanity.com/paper/2203.09093/) (2022)
- [Efficient Few-Shot Object Detection via Knowledge Inheritance](https://mlsanity.com/paper/2203.12224/) (2022)
- [Breaking Immutable: Information-Coupled Prototype Elaboration for Few-Shot Object Detection](https://mlsanity.com/paper/2211.14782/) (2022)
- [Reference Twice: A Simple and Unified Baseline for Few-Shot Instance Segmentation](https://mlsanity.com/paper/2301.01156/) (2023)
- [Multi-Faceted Distillation of Base-Novel Commonality for Few-shot Object Detection](https://mlsanity.com/paper/2207.11184/) (2022)
- [A Comparative Review of Recent Few-Shot Object Detection Algorithms](https://mlsanity.com/paper/2111.00201/) (2021)
- [Towards Efficient Use of Multi-Scale Features in Transformer-Based Object Detectors](https://mlsanity.com/paper/2208.11356/) (2022)
- [NIFF: Alleviating Forgetting in Generalized Few-Shot Object Detection via Neural Instance Feature Forging](https://mlsanity.com/paper/2303.04958/) (2023)
- [The Art of Prompting: Event Detection based on Type Specific Prompts](https://mlsanity.com/paper/2204.07241/) (2022)
- [Integrally Migrating Pre-trained Transformer Encoder-decoders for Visual Object Detection](https://mlsanity.com/paper/2205.09613/) (2022)
- [TextManiA: Enriching Visual Feature by Text-driven Manifold Augmentation](https://mlsanity.com/paper/2307.14611/) (2023)
- [Mobile Robot Manipulation using Pure Object Detection](https://mlsanity.com/paper/2201.12437/) (2022)
- [Recent Few-Shot Object Detection Algorithms: A Survey with Performance Comparison](https://mlsanity.com/paper/2203.14205/) (2022)

## Abstract (arXiv)

Few-shot object detection has made substantial progressby representing novel class objects using the feature representation learned upon a set of base class objects. However,an implicit contradiction between novel class classification and representation is unfortunately ignored. On the one hand, to achieve accurate novel class classification, the distributions of either two base classes must be far away fromeach other (max-margin). On the other hand, to precisely represent novel classes, the distributions of base classes should be close to each other to reduce the intra-class distance of novel classes (min-margin). In this paper, we propose a class margin equilibrium (CME) approach, with the aim to optimize both feature space partition and novel class reconstruction in a systematic way. CME first converts the few-shot detection problem to the few-shot classification problem by using a fully connected layer to decouple localization features. CME then reserves adequate margin space for novel classes by introducing simple-yet-effective class margin loss during feature learning. Finally, CME pursues margin equilibrium by disturbing the features of novel class instances in an adversarial min-max fashion. Experiments on Pascal VOC and MS-COCO datasets show that CME significantly improves upon two baseline detectors (up to 3~ 5% in average), achieving state-of-the-art performance. Code is available at https://github.com/Bohao-Lee/CME .