# FeatureNMS: Non-Maximum Suppression by Learning Feature Embeddings

arXiv: https://arxiv.org/abs/2002.07662 · HTML: https://mlsanity.com/paper/2002.07662/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 Feb 2020.

## Built on by (most cited first, 12 of 12)

- [Progressive End-to-End Object Detection in Crowded Scenes](https://mlsanity.com/paper/2203.07669/) (2022)
- [ODAM: Gradient-based instance-specific visual explanations for object detection](https://mlsanity.com/paper/2304.06354/) (2023)
- [A Machine Learning Surrogate Modeling Benchmark for Temperature Field Reconstruction of Heat-Source Systems](https://mlsanity.com/paper/2108.08298/) (2021)
- [PlaneSegNet: Fast and Robust Plane Estimation Using a Single-stage Instance Segmentation CNN](https://mlsanity.com/paper/2103.15428/) (2021)
- [Fuzzy-NMS: Improving 3D Object Detection with Fuzzy Classification in NMS](https://mlsanity.com/paper/2310.13951/) (2023)
- [Differentiable NMS via Sinkhorn Matching for End-to-End Fabric Defect Detection](https://mlsanity.com/paper/2505.07040/) (2025)
- [PSRR-MaxpoolNMS: Pyramid Shifted MaxpoolNMS with Relationship Recovery](https://mlsanity.com/paper/2105.12990/) (2021)
- [From Algorithm to Hardware: A Survey on Efficient and Safe Deployment of Deep Neural Networks](https://mlsanity.com/paper/2405.06038/) (2024)
- [Object Tracking by Detection with Visual and Motion Cues](https://mlsanity.com/paper/2101.07549/) (2021)
- [Work-Efficient Parallel Non-Maximum Suppression Kernels](https://mlsanity.com/paper/2502.00535/) (2025)
- [Density-based Object Detection in Crowded Scenes](https://mlsanity.com/paper/2504.09819/) (2025)
- [The Label Imitation Game: Turing Test Network for Zero-Shot Pseudo-Label Pruning](https://mlsanity.com/paper/2606.30875/) (2026)

## Abstract (arXiv)

Most state of the art object detectors output multiple detections per object. The duplicates are removed in a post-processing step called Non-Maximum Suppression. Classical Non-Maximum Suppression has shortcomings in scenes that contain objects with high overlap: This heuristic assumes that a high overlap between two bounding boxes corresponds to a high probability of one being a duplicate. We propose FeatureNMS to solve this problem. FeatureNMS recognizes duplicates not only based on the intersection over union between the bounding boxes, but also based on the difference of feature vectors. These feature vectors can encode more information like visual appearance. Our approach outperforms classical NMS and derived approaches and achieves state of the art performance.