The smart city reference pipeline shows how to integrate various media building blocks, with analytics powered by the OpenVINO™ Toolkit, for traffic or stadium sensing, analytics and management tasks.
BSD 3-Clause "New" or "Revised" License
192
stars
83
forks
source link
add new object detect pipeline with yolo-v4-tiny-tf (FP32) enable #829
1- add a new pipeline with yolo-v4-tiny-tf (FP32) enabled, and the pipeline version is 5 for RTSP.
2- enable parameter PIPELINE_VERSION for pipeline control, the default value is 1 for traffic and 2 for stadium, to enable yolov4 pipeline for traffic, we need to change its value to 5 .
( the config is at deployment/kubernetes/helm/templates/analytics.yaml:62 for kubernetes )
3- update openvino to 2021.4.2 version as yolov4 won't work on openvino 2021.3
4- model yolo-v4-tiny-tf (FP32 ) will be downloaded during building.
1- add a new pipeline with yolo-v4-tiny-tf (FP32) enabled, and the pipeline version is 5 for RTSP. 2- enable parameter PIPELINE_VERSION for pipeline control, the default value is 1 for traffic and 2 for stadium, to enable yolov4 pipeline for traffic, we need to change its value to 5 . ( the config is at deployment/kubernetes/helm/templates/analytics.yaml:62 for kubernetes ) 3- update openvino to 2021.4.2 version as yolov4 won't work on openvino 2021.3 4- model yolo-v4-tiny-tf (FP32 ) will be downloaded during building.