face_detection_and_landmarks.sh
demonstrates face detection and facial landmarking on one video file source.- This is done by running a face detection pipeline (infer + postprocessing), cropping and scaling all detected faces, and sending them into the 2nd network of facial landmarking. All resulting detections and landmarks are then aggregated and drawn on the original frame. The two networks are running using one Hailo-8 device with two
hailonet
elements.
./face_detection_and_landmarks.sh [OPTIONS] [-i INPUT_PATH]
-i --input
is an optional flag, a path to the video/camera displayed.--print-gst-launch
prints the ready gst-launch command without running it--show-fps
optional - enables printing FPS on screen--max-camera-resolution
The maximun input resolution from camera as an input
cd $TAPPAS_WORKSPACE/apps/h8/gstreamer/general/cascading_networks
./face_detection_and_landmarks.sh
The output should look like:
lightface_slim
in resolution of 320X240X3 - https://github.com/hailo-ai/hailo_model_zoo/blob/master/hailo_model_zoo/cfg/networks/lightface_slim.yamltddfa_mobilenet_v1
in resolution of 120X120X3 - https://github.com/hailo-ai/hailo_model_zoo/blob/master/hailo_model_zoo/cfg/networks/tddfa_mobilenet_v1.yaml
This app is based on our cascaded networks pipeline template
object_detection_and_pose_estimation.sh
demonstrates object detection and pose estimation on one video file source.- This is done by running an object detection pipeline, cropping and scaling each detected person, and sending them into a 2nd network of pose estimation. All resulting detections and landmarks are then aggregated and drawn on the original frame. The two networks are running using one Hailo-8 device with two
hailonet
elements.
./object_detection_and_pose_estimation.sh [OPTIONS] [-i INPUT_PATH]
-i --input
is an optional flag, a path to the video/camera displayed.--print-gst-launch
prints the ready gst-launch command without running it--show-fps
optional - enables printing FPS on screen--max-camera-resolution
The maximun input resolution from camera as an input
cd $TAPPAS_WORKSPACE/apps/h8/gstreamer/general/cascading_networks
./object_detection_and_pose_estimation.sh
The output should look like:
Joined together:
- 'yolov5m_wo_spp_60p' - https://github.com/hailo-ai/hailo_model_zoo/blob/master/hailo_model_zoo/cfg/networks/yolov5m_wo_spp_60p.yaml
- 'mspn_regnetx_800mf' - https://github.com/hailo-ai/hailo_model_zoo/blob/master/hailo_model_zoo/cfg/networks/mspn_regnetx_800mf.yaml
This app is based on our cascaded networks pipeline template