Install Tensorflow API and example for Object Detection

          Hi guys, I'm going to show you how to install Tensorflow on your Windows PC. Then we will use the Object detection API as an example of object recognition.

          The software required before installation: Python 64bit + PIP




Step 1: Use PIP settings tensorflow. If you have not installed PIP then you need to install PI first.
- PIP will be located in ~ / Python35 / Scrpits / [Example:C:\Users\h2phamth\AppData\Local\Programs\Python\Python35\Scripts]
-Command:

# For CPU
pip install tensorflow
# For GPU
pip install tensorflow-gpu

     

Step 2: Download the source code model on github and the protocol code then unpack it
Link model: https://github.com/tensorflow/models/
Link protol: https://github.com/google/protobuf/releases

Step 3: Perform the installation

# Từ tensorflow/models/research/
protoc object_detection/protos/*.proto --python_out=.

Step 4: User jupyter to open exiting example.

Bước 5: Open file exiting at: (C:\tensorflow\models-master\research\object_detection\object_detection_tutorial.ipynb) then convert to format .py

Step 6: Modify some sample code to use Camera or internal video.

Video:

You can subcribe Youtube and website to follow more post.

COCO-trained models {#coco-models}: You can use best one which suitable with your program

Model name	Speed (ms)	COCO mAP[^1]	Outputs
ssd_mobilenet_v1_coco	30	21	Boxes
ssd_inception_v2_coco	42	24	Boxes
faster_rcnn_inception_v2_coco	58	28	Boxes
faster_rcnn_resnet50_coco	89	30	Boxes
faster_rcnn_resnet50_lowproposals_coco	64		Boxes
rfcn_resnet101_coco	92	30	Boxes
faster_rcnn_resnet101_coco	106	32	Boxes
faster_rcnn_resnet101_lowproposals_coco	82		Boxes
faster_rcnn_inception_resnet_v2_atrous_coco	620	37	Boxes
faster_rcnn_inception_resnet_v2_atrous_lowproposals_coco	241		Boxes
faster_rcnn_nas	1833	43	Boxes
faster_rcnn_nas_lowproposals_coco

Kitti-trained models {#kitti-models}

Model name	Speed (ms)	Pascal mAP@0.5 (ms)	Outputs
faster_rcnn_resnet101_kitti	79	87	Boxes

Open Images-trained models {#open-images-models}

Model name	Speed (ms)	Open Images mAP@0.5[^2]	Outputs
faster_rcnn_inception_resnet_v2_atrous_oid	727	37	Boxes
faster_rcnn_inception_resnet_v2_atrous_lowproposals_oid	347		Boxe