motokimura / VisDrone_PyTorch_YOLOv3

YOLOv3 in Pytorch

Pytorch implementation of YOLOv3

What's New

• 18/11/27 COCO AP results of darknet (training) are reproduced with the same training conditions
• 18/11/20 verified inference COCO AP[IoU=0.50:0.95] = 0.302 (paper: 0.310), val5k, 416x416
• 18/11/20 verified inference COCO AP[IoU=0.50] = 0.544 (paper: 0.553), val5k, 416x416

Performance

	Original (darknet)	Ours (pytorch)
COCO AP[IoU=0.50:0.95], inference	0.310	0.302
COCO AP[IoU=0.50], inference	0.553	0.544
COCO AP[IoU=0.50:0.95], training	0.310	to be updated
COCO AP[IoU=0.50], training	0.553	to be updated

We have verified that COCO val results of darknet are reproduced in the condition where only random resizing is used:

Installation

Requirements

• Python 3.6+
• Numpy (verified as operable: 1.15.2)
• OpenCV
• Matplotlib
• Pytorch (verified as operable: v0.4.0)
• Cython (verified as operable: v0.29.1)
• pycocotools (verified as operable: v2.0.0)

optional:

• tensorboard (>1.7.0)
• tensorboardX

Docker Environment

We provide a Dockerfile to build an environment that meets the above requirements.

# build docker image
$ nvidia-docker build -t yolov3-in-pytorch-image --build-arg UID=`id -u` -f docker/Dockerfile .
# create docker container and login bash
$ nvidia-docker run -it -v `pwd`:/work --name yolov3-in-pytorch-container yolov3-in-pytorch-image
docker@4d69df209f4a:/work$ python train.py --help

Download pretrained weights

download the pretrained file from the author's project page:

$ mkdir weights
$ cd weights/
$ bash ../requirements/download_weights.sh

COCO 2017 dataset:

the COCO dataset is downloaded and unzipped by:

$ bash requirements/getcoco.sh

Inference with Pretrained Weights

To detect objects in the sample image, just run:

$ python demo.py --image data/mountain.png --detect_thresh 0.5 --weights_path weights/yolov3.weights

Train

$ python train.py --help
usage: train.py [-h] [--cfg CFG] [--weights_path WEIGHTS_PATH] [--n_cpu N_CPU]
                [--checkpoint_interval CHECKPOINT_INTERVAL]
                [--eval_interval EVAL_INTERVAL] [--checkpoint CHECKPOINT]
                [--checkpoint_dir CHECKPOINT_DIR] [--use_cuda USE_CUDA]
                [--debug] [--tfboard TFBOARD]

optional arguments:
  -h, --help            show this help message and exit
  --cfg CFG             config file. see readme
  --weights_path WEIGHTS_PATH
                        darknet weights file
  --n_cpu N_CPU         number of workers
  --checkpoint_interval CHECKPOINT_INTERVAL
                        interval between saving checkpoints
  --eval_interval EVAL_INTERVAL
                        interval between evaluations
  --checkpoint CHECKPOINT
                        pytorch checkpoint file path
  --checkpoint_dir CHECKPOINT_DIR
                        directory where checkpoint files are saved
  --use_cuda USE_CUDA
  --debug               debug mode where only one image is trained
  --tfboard TFBOARD     tensorboard path for logging

example:

$ python train.py --weights_path weights/darknet53.conv.74 --tfboard log

The train configuration is written in yaml files located in config folder. We use the following format:

MODEL:
  TYPE: YOLOv3
  BACKBONE: darknet53
TRAIN:
  LR: 0.001
  MOMENTUM: 0.9
  DECAY: 0.0005
  BURN_IN: 1000 # duration (iters) for learning rate burn-in
  MAXITER: 500000
  STEPS: (400000, 450000) # lr-drop iter points
  BATCHSIZE: 4 
  SUBDIVISION: 16 # num of minibatch inner-iterations
  IMGSIZE: 608 # initial image size
  CONFWEIGHT: 1 # not used
  LOSSTYPE: l2 # loss type for w, h
  IGNORETHRE: 0.7 # IoU threshold for learning conf
  RANDRESIZE: True # enable random resizing
TEST:
  CONFTHRE: 0.8 # not used
  NMSTHRE: 0.45 # same as official darknet
  IMGSIZE: 416 # this can be changed to measure acc-speed tradeoff
NUM_GPUS: 1

Evaluate COCO AP

$ python train.py --cfg config/yolov3_eval.cfg --eval_interval 1 [--ckpt ckpt_path] [--weights_path weights_path]

TODOs

• Precision Evaluator (bbox, COCO metric)
• Modify the target builder
• Modify loss calculation
• Training Scheduler
• Weight initialization
• Augmentation : Resizing
• Augmentation : Random Distortion
• Augmentation : Jitter
• Augmentation : Flip

Paper

YOLOv3: An Incremental Improvement

Joseph Redmon, Ali Farhadi

[Paper] [Original Implementation] [Author's Project Page]

Credit

@article{yolov3,
  title={YOLOv3: An Incremental Improvement},
  author={Redmon, Joseph and Farhadi, Ali},
  journal = {arXiv},
  year={2018}
}