最近準備實踐一下 poly-yolo，所以多做一些筆記。

論文原作者的poly-yolo項目在GitLab上面Poly-YOLO。
該項目基于tensorflow1.15
在Github上面頁有一些復現(xiàn)。

1、任意多邊形檢測

介于目標檢測與實例分割之間，可用于檢測指定邊數(shù)，任意多邊形的神經網(wǎng)絡。基于Yolov5模型進行開發(fā)。
本項目基于該項目進行二次開發(fā)https://github.com/ultralytics/yolov5/
項目是基于pytorch實現(xiàn)的。
Yolo-ArbPolygon

2、poly-yolo keras

poly-yolo

3、polyp_detection_based_on_yolov3

項目基于pytorch
polyp_detection_based_on_yolov3
We apply YOLO v3 on polyp detection

The data is from CVC colon DB, 380 photos from 15 pieces of colonoscopy.

We retrained the network for 1000 epoch with batch size of 16, 80% of data for training and 20% for testing.

The result is as shown, and the highest one reached the 96.9% of mAP.

The weight is on https://drive.google.com/file/d/1R1rT2CjVm6U1NijyHUgef7aUpLbgI86l/view?usp=sharing, use detect.py to detect new images.

polyp_detection_based_on_yolov3

4、poly-YOLOv3

poly-YOLOv3
項目基于pytorch
poly-YOLOv3
An arbitrary 4-points polygons detector based on YOLOv3

Work in progress, might never be finished !

5、Poly-YOLO

Poly-YOLO
項目基于pytorch實現(xiàn)
Poly-YOLO
This repo contains pytorch implementation of CVPR2020 Poly-YOLO: higher speed, more precise detection and instance segmentation for YOLOv3.

6、yolov3-polygon

yolov3-polygon
項目基于pytorch實現(xiàn)
This is a implementation of rotation object detecion based on YOLOv3-quadrangle. I upgraded it to support pytorch 1.1 or higher and fix some bugs. Object detection in arbitrary orientations is achieved by detecting four corner points, the model has been tested on remote sensing dataset UCAS-AOD. The results and trained models can be found here.

總結

以上是生活随笔為你收集整理的poly-yolo笔记的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網(wǎng)站內容還不錯，歡迎將生活随笔推薦給好友。