# Poly-YOLO: higher speed, more precise detection and instance segmentation for YOLOv3

arXiv: https://arxiv.org/abs/2005.13243 · HTML: https://mlsanity.com/paper/2005.13243/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 May 2020.

## Built on by (most cited first, 22 of 22)

- [OmniDet: Surround View Cameras based Multi-task Visual Perception Network for Autonomous Driving](https://mlsanity.com/paper/2102.07448/) (2021)
- [Real-time Instance Segmentation of Surgical Instruments using Attention and Multi-scale Feature Fusion](https://mlsanity.com/paper/2111.04911/) (2021)
- [OAM-TCD: A globally diverse dataset of high-resolution tree cover maps](https://mlsanity.com/paper/2407.11743/) (2024)
- [Few-Shot Learning for Road Object Detection](https://mlsanity.com/paper/2101.12543/) (2021)
- [YOLO-ReT: Towards High Accuracy Real-time Object Detection on Edge GPUs](https://mlsanity.com/paper/2110.13713/) (2021)
- [Runtime Monitoring for Out-of-Distribution Detection in Object Detection Neural Networks](https://mlsanity.com/paper/2212.07773/) (2022)
- [SCoRe: Submodular Combinatorial Representation Learning](https://mlsanity.com/paper/2310.00165/) (2023)
- [YOLOv1 to YOLOv10: The fastest and most accurate real-time object detection systems](https://mlsanity.com/paper/2408.09332/) (2024)
- [CenterPoly: real-time instance segmentation using bounding polygons](https://mlsanity.com/paper/2108.08923/) (2021)
- [Surround-View Cameras based Holistic Visual Perception for Automated Driving](https://mlsanity.com/paper/2206.05542/) (2022)
- [FisheyeDetNet: 360deg Surround view Fisheye Camera based Object Detection System for Autonomous Driving](https://mlsanity.com/paper/2404.13443/) (2024)
- [DPPD: Deformable Polar Polygon Object Detection](https://mlsanity.com/paper/2304.02250/) (2023)
- [Real-time instance segmentation with polygons using an Intersection-over-Union loss](https://mlsanity.com/paper/2305.05490/) (2023)
- [Streaming Object Detection on Fisheye Cameras for Automatic Parking](https://mlsanity.com/paper/2305.14713/) (2023)
- [DenseSeg: Joint Learning for Semantic Segmentation and Landmark Detection Using Dense Image-to-Shape Representation](https://mlsanity.com/paper/2405.19746/) (2024)
- [Mirror-Yolo: A Novel Attention Focus, Instance Segmentation and Mirror Detection Model](https://mlsanity.com/paper/2202.08498/) (2022)
- [Semi-supervised object detection based on single-stage detector for thighbone fracture localization](https://mlsanity.com/paper/2210.10998/) (2022)
- [Marine Debris Detection in Satellite Surveillance using Attention Mechanisms](https://mlsanity.com/paper/2307.04128/) (2023)
- [CenterDisks: Real-time instance segmentation with disk covering](https://mlsanity.com/paper/2403.03296/) (2024)
- [A transition towards virtual representations of visual scenes](https://mlsanity.com/paper/2410.07987/) (2024)
- [Dual-Stream Attention with Multi-Modal Queries for Object Detection in Transportation Applications](https://mlsanity.com/paper/2508.04868/) (2025)
- [ReLeaf: Benchmarking Leaf Segmentation across Domains and Species](https://mlsanity.com/paper/2605.03784/) (2026)

## Abstract (arXiv)

We present a new version of YOLO with better performance and extended with instance segmentation called Poly-YOLO. Poly-YOLO builds on the original ideas of YOLOv3 and removes two of its weaknesses: a large amount of rewritten labels and inefficient distribution of anchors. Poly-YOLO reduces the issues by aggregating features from a light SE-Darknet-53 backbone with a hypercolumn technique, using stairstep upsampling, and produces a single scale output with high resolution. In comparison with YOLOv3, Poly-YOLO has only 60% of its trainable parameters but improves mAP by a relative 40%. We also present Poly-YOLO lite with fewer parameters and a lower output resolution. It has the same precision as YOLOv3, but it is three times smaller and twice as fast, thus suitable for embedded devices. Finally, Poly-YOLO performs instance segmentation using bounding polygons. The network is trained to detect size-independent polygons defined on a polar grid. Vertices of each polygon are being predicted with their confidence, and therefore Poly-YOLO produces polygons with a varying number of vertices.