Update

Recently, I optimize my YOLO project:

https://github.com/yjh0410/PyTorch_YOLO-Family

In my new YOLO project, you can enjoy:

• a new and stronger YOLOv1
• a new and stronger YOLOv2
• YOLOv3
• YOLOv3 with SPP
• YOLOv3 with DilatedEncoder
• YOLOv4-Exp ~ (I'm trying to make it better)
• YOLO-Tiny
• YOLO-Nano

This project

In this project, you can enjoy:

• YOLOv2 with DarkNet-19
• YOLOv2 with ResNet-50
• YOLOv2Slim
• YOLOv3
• YOLOv3Spp
• YOLOv3Tiny

I just want to provide a good YOLO project for everyone who is interested in Object Detection.

Weights

Google Drive: https://drive.google.com/drive/folders/1T5hHyGICbFSdu6u2_vqvxn_puotvPsbd?usp=sharing

BaiDuYunDisk: https://pan.baidu.com/s/1tSylvzOVFReUAvaAxKRSwg Password d266

You can download all my models from the above links.

DarkNet-19, DarkNet-53, CSPDarkNet-53

First of all, I reproduce the backbones used in YOLO.

DarkNet-19

	size	top1
darknet19	224	72.96
darknet19	448	75.52

DarkNet-53

	size	top1
darknet53	224	75.42
darknet53	448	77.76

CSPDarkNet-53

	size	top1
cspdarknet53	224	75.7
cspdarknet53-448	448	76.9

YOLOv2

YOLOv2 with DarkNet-19

Tricks

Tricks in official paper:

• batch norm
• hi-res classifier
• convolutional
• anchor boxes
• new network
• dimension priors
• location prediction
• passthrough
• multi-scale
• hi-red detector

Then I train my YOLOv2 on VOC dataset. I really enjoy yolo. It is so amazing! I think I have achieved this goal:

	size	Original (darknet)	Ours (pytorch) 160peochs	Ours (pytorch) 250epochs
VOC07 test	416	76.8	76.0	77.1
VOC07 test	544	78.6	77.0	78.1

With 160 training epochs, my yolo-v2 only gets 76.0 mAP with 416 input size and 77.0 mAP with 544 input size. To be better, I add another 90 epochs. With 250 training epochs, my yolo-v2 performs very well !

During testing stage, I set conf thresh as 0.001 and set nms thresh as 0.5 to obtain above results. To make my model faster, I set conf thresh as 0.01. With this higher conf thresh, my yolo-v2 still performs very well and gets 77.0 mAP with 416 input size and 78.0 mAP with 544 input size.

I visualize some detection results whose score is over 0.3 on VOC 2007 test:

COCO:

	data	AP	AP50	AP75	AP_S	AP_M	AP_L
Original (darknet)	COCO test-dev	21.6	44.0	19.2	5.0	22.4	35.5
Ours (pytorch)	COCO test-dev	26.8	46.6	26.8	5.8	27.4	45.2
Ours (pytorch)	COCO eval	26.6	46.0	26.7	5.9	27.8	47.1

In TITAN Xp, my yolo-v2 runs at 100+ FPS, so it's very fast. I have no any TITAN X GPU, and I can't run my model in a X GPU. Sorry, guys~

Before I tell you how to use this project, I must say one important thing about difference between origin yolo-v2 and mine:

• For data augmentation, I copy the augmentation codes from the https://github.com/amdegroot/ssd.pytorch which is a superb project reproducing the SSD. If anyone is interested in SSD, just clone it to learn !(Don't forget to star it !)

So I don't write data augmentation by myself. I'm a little lazy~~

My loss function and groundtruth creator both in the tools.py, and you can try to change any parameters to improve the model.

YOLOv2 with ResNet-50

I replace darknet-19 with resnet-50 and get a better result on COCO-val

	data	AP	AP50	AP75	AP_S	AP_M	AP_L
Our YOLOv2-320	COCO eval	25.8	44.6	25.9	4.6	26.8	47.9
Our YOLOv2-416	COCO eval	29.0	48.8	29.7	7.4	31.9	48.3
Our YOLOv2-512	COCO eval	30.4	51.6	30.9	10.1	34.9	46.6
Our YOLOv2-544	COCO eval	30.4	51.9	30.9	11.1	35.8	45.5
Our YOLOv2-608	COCO eval	29.2	51.6	29.1	13.6	36.8	40.5

YOLOv3

I train my YOLOv3 on VOC:

	size	Original (darknet)	Ours (pytorch) 250epochs
VOC07 test	416	80.25	81.4

I use the same training strategy to my YOLOv2. My data-processing code is a little different from official YOLOv3. For more details, you can check my code files.

COCO:

Original YOLOv3:

	data	AP	AP50	AP75	AP_S	AP_M	AP_L
YOLOv3-320	COCO test-dev	28.2	51.5	-	-	-	-
YOLOv3-416	COCO test-dev	31.0	55.3	-	-	-	-
YOLOv3-608	COCO test-dev	33.0	57.0	34.4	18.3	35.4	41.9

Our YOLOv3_PyTorch:

	data	AP	AP50	AP75	AP_S	AP_M	AP_L
YOLOv3-320	COCO test-dev	33.1	54.1	34.5	12.1	34.5	49.6
YOLOv3-416	COCO test-dev	36.0	57.4	37.0	16.3	37.5	51.1
YOLOv3-608	COCO test-dev	37.6	59.4	39.9	20.4	39.9	48.2

My YOLOv3 is very stronger and better, right?

I also visualize some detection results whose score is over 0.3 on COCO 2017-val:

So, just have fun !

YOLOv3SPP

On COCO:

	data	AP	AP50	AP75	AP_S	AP_M	AP_L
YOLOv3Spp-320	COCO eval	32.78	53.79	33.9	12.4	35.5	50.6
YOLOv3Spp-416	COCO eval	35.66	57.09	37.4	16.8	38.1	50.7
YOLOv3Spp-608	COCO eval	37.52	59.44	39.3	21.5	40.6	49.6

My YOLOv3SPP is not better than my YOLOv3. It is a little confusing ...

YOLOv2Slim

I build a very simple lightweight backbone: darknet_tiny

I replace the darknet19 used in YOLOv2 with darknet_tiny.

My SlimYOLOv2 is fast and strong. On VOC, it gets 70.7 mAP and 100+ FPS on 1660ti GPU.

Just enjoy it.

And, I'm still trying to make it faster without too much drop of precision.

YOLOv3Tiny

We evaluate our YOLOv3Tiny on COCO-val with inputsize 608:

	data	AP	AP50	AP75	AP_S	AP_M	AP_L
(official) YOLOv3Tiny	COCO test-dev	-	33.1	-	-	-	-
(Our) YOLOv3Tiny	COCO val	15.9	33.8	12.8	7.6	17.7	22.4

Installation

• Pytorch-gpu 1.1.0/1.2.0/1.3.0
• Tensorboard 1.14.
• opencv-python, python3.6/3.7

Dataset

VOC Dataset

I copy the download files from the following excellent project: https://github.com/amdegroot/ssd.pytorch

I have uploaded the VOC2007 and VOC2012 to BaiDuYunDisk, so for researchers in China, you can download them from BaiDuYunDisk:

Link：https://pan.baidu.com/s/1tYPGCYGyC0wjpC97H-zzMQ

Password：4la9

You will get a VOCdevkit.zip, then what you need to do is just to unzip it and put it into data/. After that, the whole path to VOC dataset is data/VOCdevkit/VOC2007 and data/VOCdevkit/VOC2012.

Download VOC2007 trainval & test

# specify a directory for dataset to be downloaded into, else default is ~/data/
sh data/scripts/VOC2007.sh # <directory>

Download VOC2012 trainval

# specify a directory for dataset to be downloaded into, else default is ~/data/
sh data/scripts/VOC2012.sh # <directory>

MSCOCO Dataset

I copy the download files from the following excellent project: https://github.com/DeNA/PyTorch_YOLOv3

Download MSCOCO 2017 dataset

Just run sh data/scripts/COCO2017.sh. You will get COCO train2017, val2017, test2017.

Train

VOC

python train.py -d voc --cuda -v [select a model] -hr -ms --ema

You can run python train.py -h to check all optional argument.

COCO

If you have only one gpu:

python train.py -d coco --cuda -v [select a model] -hr -ms --ema

If you have multi gpus like 8, and you put 4 images on each gpu:

python -m torch.distributed.launch --nproc_per_node=8 train.py -d coco --cuda -v [select a model] -hr -ms --ema \
                                                                        -dist \
                                                                        --sybn \
                                                                        --num_gpu 8\
                                                                        --batch_size 4

Test

VOC

python test.py -d voc --cuda -v [select a model] --trained_model [ Please input the path to model dir. ]

COCO

python test.py -d coco-val --cuda -v [select a model] --trained_model [ Please input the path to model dir. ]

Evaluation

VOC

python eval.py -d voc --cuda -v [select a model] --train_model [ Please input the path to model dir. ]

COCO

To run on COCO_val:

python eval.py -d coco-val --cuda -v [select a model] --train_model [ Please input the path to model dir. ]

To run on COCO_test-dev(You must be sure that you have downloaded test2017):

python eval.py -d coco-test --cuda -v [select a model] --train_model [ Please input the path to model dir. ]

You will get a .json file which can be evaluated on COCO test server.