FastDeploy/examples/vision/detection/paddledetection

English | 简体中文

PaddleDetection Model Deployment

Model Description

• PaddleDetection Release/2.4

List of Supported Models

Now FastDeploy supports the deployment of the following models

• PP-YOLOE(including PP-YOLOE+) models
• PicoDet models
• PP-YOLO models(including v2)
• YOLOv3 models
• YOLOX models
• FasterRCNN models
• MaskRCNN models
• SSD models
• YOLOv5 models
• YOLOv6 models
• YOLOv7 models
• RTMDet models

Export Deployment Model

Before deployment, PaddleDetection needs to be exported into the deployment model. Refer to Export Models for more details.

Attention

• Do not perform NMS removal when exporting the model
• If you are running a native TensorRT backend (not a Paddle Inference backend), do not add the --trt parameter
• Do not add the parameter fuse_normalize=True when exporting the model

Download Pre-trained Model

For developers' testing, models exported by PaddleDetection are provided below. Developers can download them directly.

The accuracy metric is from model descriptions in PaddleDetection. Refer to them for details.

Model	Parameter Size	Accuracy	Note
picodet_l_320_coco_lcnet	23MB	Box AP 42.6%
ppyoloe_crn_l_300e_coco	200MB	Box AP 51.4%
ppyoloe_plus_crn_m_80e_coco	83.3MB	Box AP 49.8%
ppyolo_r50vd_dcn_1x_coco	180MB	Box AP 44.8%	TensorRT not supported yet
ppyolov2_r101vd_dcn_365e_coco	282MB	Box AP 49.7%	TensorRT not supported yet
yolov3_darknet53_270e_coco	237MB	Box AP 39.1%
yolox_s_300e_coco	35MB	Box AP 40.4%
faster_rcnn_r50_vd_fpn_2x_coco	160MB	Box AP 40.8%	TensorRT not supported yet
mask_rcnn_r50_1x_coco	128M	Box AP 37.4%, Mask AP 32.8%	TensorRT、ORT not supported yet
ssd_mobilenet_v1_300_120e_voc	24.9M	Box AP 73.8%	TensorRT、ORT not supported yet
ssd_vgg16_300_240e_voc	106.5M	Box AP 77.8%	TensorRT、ORT not supported yet
ssdlite_mobilenet_v1_300_coco	29.1M		TensorRT、ORT not supported yet
rtmdet_l_300e_coco	224M	Box AP 51.2%
rtmdet_s_300e_coco	42M	Box AP 44.5%
yolov5_l_300e_coco	183M	Box AP 48.9%
yolov5_s_300e_coco	31M	Box AP 37.6%
yolov6_l_300e_coco	229M	Box AP 51.0%
yolov6_s_400e_coco	68M	Box AP 43.4%
yolov7_l_300e_coco	145M	Box AP 51.0%
yolov7_x_300e_coco	277M	Box AP 53.0%

Detailed Deployment Documents

• Python Deployment
• C++ Deployment