English | [简体中文](README_cn.md)

# PP-YOLO

## Table of Contents
- [Introduction](#Introduction)
- [Model Zoo](#Model_Zoo)
- [Getting Start](#Getting_Start)
- [Future Work](#Future_Work)
- [Appendix](#Appendix)

## Introduction

[PP-YOLO](https://arxiv.org/abs/2007.12099) is a optimized model based on YOLOv3 in PaddleDetection，whose performance(mAP on COCO) and inference spped are better than [YOLOv4](https://arxiv.org/abs/2004.10934)，PaddlePaddle 2.0.0rc1(available on pip now) or [Daily Version](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/install/Tables.html#whl-release) is required to run this PP-YOLO。

PP-YOLO reached mmAP(IoU=0.5:0.95) as 45.9% on COCO test-dev2017 dataset, and inference speed of FP32 on single V100 is 72.9 FPS, inference speed of FP16 with TensorRT on single V100 is 155.6 FPS.

<div align="center">
  <img src="../../../docs/images/ppyolo_map_fps.png" width=500 />
</div>

PP-YOLO improved performance and speed of YOLOv3 with following methods:

- Better backbone: ResNet50vd-DCN
- Larger training batch size: 8 GPUs and mini-batch size as 24 on each GPU
- [Drop Block](https://arxiv.org/abs/1810.12890)
- [Exponential Moving Average](https://www.investopedia.com/terms/e/ema.asp)
- [IoU Loss](https://arxiv.org/pdf/1902.09630.pdf)
- [Grid Sensitive](https://arxiv.org/abs/2004.10934)
- [Matrix NMS](https://arxiv.org/pdf/2003.10152.pdf)
- [CoordConv](https://arxiv.org/abs/1807.03247)
- [Spatial Pyramid Pooling](https://arxiv.org/abs/1406.4729)
- Better ImageNet pretrain weights

## Model Zoo

### PP-YOLO

|          Model           | GPU number | images/GPU |  backbone  | input shape | Box AP<sup>val</sup> | Box AP<sup>test</sup> | V100 FP32(FPS) | V100 TensorRT FP16(FPS) | download | config  |
|:------------------------:|:----------:|:----------:|:----------:| :----------:| :------------------: | :-------------------: | :------------: | :---------------------: | :------: | :-----: |
| PP-YOLO                  |     8      |     24     | ResNet50vd |     608     |         44.8         |         45.2          |      72.9      |          155.6          | [model](https://paddlemodels.bj.bcebos.com/object_detection/dygraph/ppyolo_r50vd_dcn_1x_coco.pdparams) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/master/dygraph/configs/ppyolo/ppyolo_r50vd_dcn_1x_coco.yml)                   |
| PP-YOLO_2x               |     8      |     24     | ResNet50vd |     608     |         45.3         |         45.9          |      72.9      |          155.6          | [model](https://paddlemodels.bj.bcebos.com/object_detection/dygraph/ppyolo_r50vd_dcn_2x_coco.pdparams) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/master/dygraph/configs/ppyolo/ppyolo_r50vd_dcn_2x_coco.yml)                   |

**Notes:**

- PP-YOLO is trained on COCO train2017 dataset and evaluated on val2017 & test-dev2017 dataset，Box AP<sup>test</sup> is evaluation results of `mAP(IoU=0.5:0.95)`.
- PP-YOLO used 8 GPUs for training and mini-batch size as 24 on each GPU, if GPU number and mini-batch size is changed, learning rate and iteration times should be adjusted according [FAQ](../../../docs/FAQ.md).
- PP-YOLO inference speed is tesed on single Tesla V100 with batch size as 1, CUDA 10.2, CUDNN 7.5.1, TensorRT 5.1.2.2 in TensorRT mode.
- PP-YOLO FP32 inference speed testing uses inference model exported by `tools/export_model.py` and benchmarked by running `depoly/python/infer.py` with `--run_benchmark`. All testing results do not contains the time cost of data reading and post-processing(NMS), which is same as [YOLOv4(AlexyAB)](https://github.com/AlexeyAB/darknet) in testing method.
- TensorRT FP16 inference speed testing exclude the time cost of bounding-box decoding(`yolo_box`) part comparing with FP32 testing above, which means that data reading, bounding-box decoding and post-processing(NMS) is excluded(test method same as [YOLOv4(AlexyAB)](https://github.com/AlexeyAB/darknet) too)

## Getting Start

### 1. Training

Training PP-YOLO on 8 GPUs with following command(all commands should be run under PaddleDetection dygraph directory as default)

```bash
python -m paddle.distributed.launch --log_dir=./ppyolo_dygraph/ --gpus 0,1,2,3,4,5,6,7 tools/train.py -c configs/ppyolo/ppyolo_r50vd_dcn_1x_coco.yml &>ppyolo_dygraph.log 2>&1 &
```

### 2. Evaluation

Evaluating PP-YOLO on COCO val2017 dataset in single GPU with following commands:

```bash
# use weights released in PaddleDetection model zoo
CUDA_VISIBLE_DEVICES=0 python tools/eval.py -c configs/ppyolo/ppyolo_r50vd_dcn_1x_coco.yml -o weights=https://paddlemodels.bj.bcebos.com/object_detection/dygraph/ppyolo_r50vd_dcn_1x_coco.pdparams

# use saved checkpoint in training
CUDA_VISIBLE_DEVICES=0 python tools/eval.py -c configs/ppyolo/ppyolo_r50vd_dcn_1x_coco.yml -o weights=output/ppyolo_r50vd_dcn_1x_coco/model_final
```

For evaluation on COCO test-dev2017 dataset, `configs/ppyolo/ppyolo_test.yml` should be used, please download COCO test-dev2017 dataset from [COCO dataset download](https://cocodataset.org/#download) and decompress to pathes configured by `EvalReader.dataset` in `configs/ppyolo/ppyolo_test.yml` and run evaluation by following command:

```bash
# use weights released in PaddleDetection model zoo
CUDA_VISIBLE_DEVICES=0 python tools/eval.py -c configs/ppyolo/ppyolo_test.yml -o weights=https://paddlemodels.bj.bcebos.com/object_detection/dygraph/ppyolo_r50vd_dcn_1x_coco.pdparams

# use saved checkpoint in training
CUDA_VISIBLE_DEVICES=0 python tools/eval.py -c configs/ppyolo/ppyolo_test.yml -o weights=output/ppyolo_r50vd_dcn_1x_coco/model_final
```

Evaluation results will be saved in `bbox.json`, compress it into a `zip` package and upload to [COCO dataset evaluation](https://competitions.codalab.org/competitions/20794#participate) to evaluate.

**NOTE:** `configs/ppyolo/ppyolo_test.yml` is only used for evaluation on COCO test-dev2017 dataset, could not be used for training or COCO val2017 dataset evaluating.

### 3. Inference

Inference images in single GPU with following commands, use `--infer_img` to inference a single image and `--infer_dir` to inference all images in the directory.

```bash
# inference single image
CUDA_VISIBLE_DEVICES=0 python tools/infer.py configs/ppyolo/ppyolo_r50vd_dcn_1x_coco.yml -o weights=https://paddlemodels.bj.bcebos.com/object_detection/dygraph/ppyolo_r50vd_dcn_1x_coco.pdparams --infer_img=../demo/000000014439_640x640.jpg

# inference all images in the directory
CUDA_VISIBLE_DEVICES=0 python tools/infer.py configs/ppyolo/ppyolo_r50vd_dcn_1x_coco.yml -o weights=https://paddlemodels.bj.bcebos.com/object_detection/dygraph/ppyolo_r50vd_dcn_1x_coco.pdparams --infer_dir=../demo
```

### 4. Inferece deployment and benchmark

For inference deployment or benchmard, model exported with `tools/export_model.py` should be used and perform inference with Paddle inference library with following commands:

```bash
# export model, model will be save in output/ppyolo as default
python tools/export_model.py -c configs/ppyolo/ppyolo_r50vd_dcn_1x_coco.yml -o weights=https://paddlemodels.bj.bcebos.com/object_detection/dygraph/ppyolo_r50vd_dcn_1x_coco.pdparams

# inference with Paddle Inference library
CUDA_VISIBLE_DEVICES=0 python deploy/python/infer.py --model_dir=output_inference/ppyolo_r50vd_dcn_1x_coco --image_file=../demo/000000014439_640x640.jpg --use_gpu=True
```

Benchmark testing for PP-YOLO uses model without data reading and post-processing(NMS), export model with `--exclude_nms` to prunce NMS for benchmark testing from mode with following commands:

```bash
# export model, --exclude_nms to prune NMS part, model will be save in output/ppyolo as default
python tools/export_model.py -c configs/ppyolo/ppyolo_r50vd_dcn_1x_coco.yml -o weights=https://paddlemodels.bj.bcebos.com/object_detection/dygraph/ppyolo_r50vd_dcn_1x_coco.pdparams --exclude_nms

# FP32 benchmark
CUDA_VISIBLE_DEVICES=0 python deploy/python/infer.py --model_dir=output_inference/ppyolo_r50vd_dcn_1x_coco --image_file=../demo/000000014439_640x640.jpg --use_gpu=True --run_benchmark=True

# TensorRT FP16 benchmark
CUDA_VISIBLE_DEVICES=0 python deploy/python/infer.py --model_dir=output_inference/ppyolo_r50vd_dcn_1x_coco --image_file=../demo/000000014439_640x640.jpg --use_gpu=True --run_benchmark=True --run_mode=trt_fp16
```

## Future work

1. more PP-YOLO tiny model
2. PP-YOLO model with more backbones

## Appendix

Optimizing method and ablation experiments of PP-YOLO compared with YOLOv3.

| NO.  |        Model                 | Box AP<sup>val</sup> | Box AP<sup>test</sup> | Params(M) | FLOPs(G) | V100 FP32 FPS |
| :--: | :--------------------------- | :------------------: |:--------------------: | :-------: | :------: | :-----------: |
|  A   | YOLOv3-DarkNet53             |         38.9         |           -           |   59.13   |  65.52   |      58.2     |
|  B   | YOLOv3-ResNet50vd-DCN        |         39.1         |           -           |   43.89   |  44.71   |      79.2     |
|  C   | B + LB + EMA + DropBlock     |         41.4         |           -           |   43.89   |  44.71   |      79.2     |
|  D   | C + IoU Loss                 |         41.9         |           -           |   43.89   |  44.71   |      79.2     |
|  E   | D + IoU Aware                |         42.5         |           -           |   43.90   |  44.71   |      74.9     |
|  F   | E + Grid Sensitive           |         42.8         |           -           |   43.90   |  44.71   |      74.8     |
|  G   | F + Matrix NMS               |         43.5         |           -           |   43.90   |  44.71   |      74.8     |
|  H   | G + CoordConv                |         44.0         |           -           |   43.93   |  44.76   |      74.1     |
|  I   | H + SPP                      |         44.3         |         45.2          |   44.93   |  45.12   |      72.9     |
|  J   | I + Better ImageNet Pretrain |         44.8         |         45.2          |   44.93   |  45.12   |      72.9     |
|  K   | J + 2x Scheduler             |         45.3         |         45.9          |   44.93   |  45.12   |      72.9     |

**Notes:**

- Performance and inference spedd are measure with input shape as 608
- All models are trained on COCO train2017 datast and evaluated on val2017 & test-dev2017 dataset，`Box AP` is evaluation results as `mAP(IoU=0.5:0.95)`.
- Inference speed is tested on single Tesla V100 with batch size as 1 following test method and environment configuration in benchmark above.
- [YOLOv3-DarkNet53](../yolov3/yolov3_darknet53_270e_coco.yml) with mAP as 39.0 is optimized YOLOv3 model in PaddleDetection，see [Model Zoo](../../../docs/MODEL_ZOO.md) for details.