This is Broiler Weight Prediction Program which is based on 3D Computer Vision. This program receives RGB data and depth data as input and outputs the average weight of broilers detected in the video.
| Python | pytorch | Windows | Mac | Linux |
|---|---|---|---|---|
| 3.8.0+ | 1.10.0+ | X | X | Ubuntu 18.04 |
Clone repo and install requirements.txt in a Python>=3.8.0 environment, including PyTorch>=1.10.0.
(tested environment cuda 11.3, cudnn 8.2.1 for cuda 11.x)
pip install torch==1.12.1+cu113 torchvision==0.13.1+cu113 --extra-index-url https://download.pytorch.org/whl/cu113
pip install -U openmim
mim install mmcv-full
git clone https://github.com/dohyeonYoon/BRODY # clone
cd BRODY
pip install -r requirements.txt # dependency install
cd BRODY/src/mmdetection/source
pip install -v -e . # mmdet installYou can inference with your own custom RGB, PGM file in /BRODY/src/input/rgb, /BRODY/src/input/depth folder.
cd src
python pyBRODY.py
Please place the downloaded weights and dataset directories in /BRODY/src/mmdetection/ directory respectively. (The download of the checkpoint file is temporarily unavailable)
weight # Mask-RCNN Weights
dataset # Datasets