Skip to content

poornima2605/Plant-Parsing

Repository files navigation

Introduction

Implementation of Quality-Aware Network for Plant Parsing

In this repository, we release the QANet code in Pytorch.

  • QANet architecture:

Installation

  • 1 x RTX GPU
  • pytorch1.6
  • python3.6.8

Install QANet following [INSTALL.md].

Data Prepare

Please follow [DATA_PREPARE.md] to download training and evaluating data.

Results and Models

QANet On Plant parsing dataset

Backbone DOWNLOAD
ResNet50 GoogleDrive

please put the pretrained weights in QANet/weights

Training

To train a model with 1 GPUs run:

python tools/train_net_all.py --cfg cfgs/CIHP/QANet/QANet_R-50c_512x384_1x.yaml --gpu_id 0

Evaluation

single-gpu evaluation,

python tools/test_net_all.py --cfg ckpts/CIHP/QANet/QANet_R-50c_512x384_1x/QANet_R-50c_512x384_1x.yaml --gpu_id 0

License

QANet is released under the MIT license. And the code is adapted from QANet Human parsing

About

Adaptation of human parsing repository to plants. Instance and semantic segmentation implementation.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published