HFDNet

Introduction

This is our PyTorch implementation of the paper "[Hierarchical Feature Differentiation-guided Network for domain adaptation object detection].

Dataset Preparing

``bash Cityscapes → Foggy Cityscapes Cityscapes: https://www.cityscapes-dataset.com/downloads/ Images leftImg8bittrainvaltest.zip (11GB) [md5]. `Annotations gtFinetrainvaltest.zip (241MB) [md5]. Foggy CityScapes: https://www.cityscapes-dataset.com/downloads/. Images leftImg8bit_trainval_foggyDBF.zip (20GB) [md5]. Annotations are the same with CityScapes. Sim10K → Cityscapes Sim10K : https://fcav.engin.umich.edu/projects/driving-in-the-matrix Pascal VOC → Clipart1k PascalVOC(2007+2012): follow the scripts in .dataset/VOC.yaml to build VOC datasets. Clipart1k: https://github.com/naoto0804/cross-domain-detection. Public power data CPLID: https://github.com/InsulatorData/InsulatorDataSet. VPMBGI: https://github.com/phd-benel/VPMBGI. # IDID: https://ieee-dataport.org/competitions/insulator-defect-detection.

🔔 '.yaml' file is in https://drive.google.com/drive/folders/1aqQfeHzpxAiJBeITXhvD-Bw0PZnLeeP0 ```

Weight

Since github can't upload large files, we uploaded the weights of the four benchmark tasks to the Google Drive

Quick Start Examples

Install

```bash

clone the project and configure the environment.

git clone https://github.com/LiuJiaji1999/HFDNet.git

the version of ultralytics is '8.2.50'

GPU-NVIDIA GeForce RTX 3090

CPU-12th Gen Intel(R) Core(TM) i9-12900

python: 3.8.18 torch: 1.12.0+cu113 torchvision: 0.13.0+cu113 numpy: 1.22.3
```

Test & Detect

bash python val.py python detect.py

Train

```bash python train.py

nohup python train.py > /home/lenovo/data/liujiaji/powerGit/dayolo/logs/improve/c2f.log 2>&1 & tail -f /home/lenovo/data/liujiaji/powerGit/dayolo/logs/improve/c2f.log

```

Dual-input

```bash /ultralytics/models/yolo/model.py /ultralytics/models/yolo/detect/init.py /ultralytics/models/yolo/detect/udatrain.py /ultralytics/data/udabuild.py # load dataset def udabuilddataloader /ultralytics/nn/udatasks.py # update model structure /ultralytics/engine/udatrainer.py # update trainer /ultralytics/utils/daca.py # compute loss /ultralytics/engine/validator.py # loss /ultralytics/cfg/default.yaml # add weight value /ultralytics/nn/modules/head.py # head pseudo /ultralytics/utils/plotting.py # outputtotarget

```

Explanation of the file

```bash

Main scripts

train.py: script to train the model val.py: script that calculates metrics using the trained model detect.py: inference script

Other scripts

distill.py: distill script export.py: export onnx scripts gap.py: /gap getFPS.py: script to calculate model storage size, model inference time-FPS heatmap.py: script to generate heat map mainprofile.py : a script that outputs the model and the parameters and calculations for each layer of the model track.py : script to track inference ```

References

Reference code links are all HERE, thanks to the spirit of open source.