tensorflow-yolov4实施方法

tensorflow-yolov4-tflite

YOLOv4: Optimal Speed and Accuracy of Object Detection

文献链接：https://arxiv.org/abs/2004.10934

代码链接：https://github.com/AlexeyAB/darknet

摘要

有大量的特征被认为可以提高卷积神经网络（CNN）的精度。需要在大型数据集上对这些特征的组合进行实际测试，并对结果进行理论证明。某些功能只在某些模型上操作，某些问题只在某些模型上操作，或只在小规模数据集上操作；而某些功能（如批处理规范化和剩余连接）适用于大多数模型、任务和数据集。我们假设这些通用特征包括加权剩余连接（WRC）、跨阶段部分连接（CSP）、跨小批量规范化（CmBN）、自对抗训练（SAT）和Mish激活。使用了新功能：WRC、CSP、CmBN、SAT、误激活、马赛克数据增强、CmBN、DropBlock正则化和CIoU丢失，并将其中一些功能结合起来，以达到最新的结果：43.5%AP（65.7%AP50）的MS COCO数据集，在Tesla V100上以约65 FPS的实时速度。

YOLOv4 Implemented in Tensorflow 2.0. Convert YOLO v4, YOLOv3, YOLO tiny .weights to .pb, .tflite and trt format for tensorflow, tensorflow lite, tensorRT.

Download yolov4.weights file: https://drive.google.com/open?id=1cewMfusmPjYWbrnuJRuKhPMwRe_b9PaT

环境需要Prerequisites

• Tensorflow 2.1.0
• tensorflow_addons 0.9.1 (required for mish activation)

Demo

# yolov4

python detect.py --weights ./data/yolov4.weights --framework tf --size 608 --image ./data/kite.jpg

# yolov4 tflite

python detect.py --weights ./data/yolov4-int8.tflite --framework tflite --size 416 --image ./data/kite.jpg

Convert to tflite

# yolov4

python convert_tflite.py --weights ./data/yolov4.weights --output ./data/yolov4.tflite

# yolov4 quantize float16

python convert_tflite.py --weights ./data/yolov4.weights --output ./data/yolov4-fp16.tflite --quantize_mode float16

# yolov4 quantize int8

python convert_tflite.py --weights ./data/yolov4.weights --output ./data/yolov4-fp16.tflite --quantize_mode full_int8 --dataset ./coco_dataset/coco/val207.txt

Convert to TensorRT

# yolov3

python save_model.py --weights ./data/yolov3.weights --output ./checkpoints/yolov3.tf --input_size 416 --model yolov3

python convert_trt.py --weights ./checkpoints/yolov3.tf --quantize_mode float16 --output ./checkpoints/yolov3-trt-fp16-416

# yolov3-tiny

python save_model.py --weights ./data/yolov3-tiny.weights --output ./checkpoints/yolov3-tiny.tf --input_size 416 --tiny

python convert_trt.py --weights ./checkpoints/yolov3-tiny.tf --quantize_mode float16 --output ./checkpoints/yolov3-tiny-trt-fp16-416

# yolov4

python save_model.py --weights ./data/yolov4.weights --output ./checkpoints/yolov4.tf --input_size 416 --model yolov4

python convert_trt.py --weights ./checkpoints/yolov4.tf --quantize_mode float16 --output ./checkpoints/yolov4-trt-fp16-416

Evaluate on COCO 2017 Dataset

# run script in /script/get_coco_dataset_2017.sh to download COCO 2017 Dataset

# preprocess coco dataset

cd data

mkdir dataset

cd ..

cd scripts

python coco_convert.py --input ./coco/annotations/instances_val2017.json --output val2017.pkl

python coco_annotation.py --coco_path ./coco

cd ..

# evaluate yolov4 model

python evaluate.py --weights ./data/yolov4.weights

cd mAP/extra

python remove_space.py

cd ..

python main.py --output results_yolov4_tf

mAP50 on COCO 2017 Dataset

 

Benchmark

python benchmarks.py --size 416 --model yolov4 --weights ./data/yolov4.weights

TensorRT performance

训练模型

# Prepare your dataset

# If you want to train from scratch:

In config.py set FISRT_STAGE_EPOCHS=0

# Run script:

python train.py

# Transfer learning:

python train.py --weights ./data/yolov4.weights

训练性能还没有完全重现，建议使用Alex的Darknet训练自己的数据，然后将.weights转换为tensorflow或tflite。