3.8 KiB
1-1-1_sample
1 video input, 1 infer task, and 1 output.
1-1-N_sample
1 video input, 1 infer task, and 2 outputs.
1-N-N_sample
1 video input and then split into 2 branches for different infer tasks, then 2 total outputs.
N-1-N_sample
2 video input and merge into 1 branch automatically for 1 infer task, then resume to 2 branches for outputs again.
N-N_sample
multi pipe exist separately and each pipe is 1-1-1 (can be any structure like 1-1-N, 1-N-N)
paddle_infer_sample
ocr based on paddle (install paddle_inference first!), 1 video input and 2 outputs (screen, rtmp)
src_des_sample
show how src nodes and des nodes work.
3 (file, rtsp, udp) input and merge into 1 infer task, then resume to 3 branches for outputs (screen, rtmp, fake)
trt_infer_sample
vehicle and plate detector based on tensorrt (install tensorrt first!), 1 video input and 3 outputs (screen, file, rtmp)
vp_logger_sample
show how vp_logger works.
face_tracking_sample
tracking for multi faces.
vehicle_tracking_sample
tracking for multi vehicles.
interaction_with_pipe_sample
show how to interact with pipe, such as start/stop channel by calling api.
record_sample
show how vp_record_node works.
message_broker_sample & message_broker_sample2
show how message broker nodes work.
mask_rcnn_sample
show image segmentation by mask-rcnn.
openpose_sample
show pose estimation by openpose network.
enet_seg_sample
show semantic segmentation by enet network.
multi_detectors_and_classifiers_sample
show multi infer node work together.
image_des_sample
show save/push image to local file or remote via udp.
image_src_sample
show read/receive image from local file or remote via udp.
rtsp_des_sample
show push video stream via rtsp, no rtsp server needed, you can visit it directly.
ba_crossline_sample
count for vehicle based on tracking, the simplest one of behaviour analysis.
plate_recognize_sample
vehicle plate detect and recognize on the whole frame (no need to detect vechile first)
vehicle_body_scan_sample
detect parts of vehicle based on side view of body
body_scan_and_plate_detect_sample
2 channels to detect parts of vehicle and detect vehicle plate, you can do something like data fusion later
app_src_sample
send data to pipeline from host coda using app_src_node
vehicle_cluster_based_on_classify_encoding_sample
vehicle cluster based on labels(classify) and encoding(feature extract), pipeline would display 3 windows (cluster by t-SNE, cluster by labels, detect result)
ba_stop_sample
vehicle stop behaviour analysis
similiarity search
flask demo for vehicle and face similiarity search
behaviour analysis
flask demo for crossline and stop
property and similiarity search
flask demo for vehicle search by similiarity and properties
ba_jam_sample
traffic jam behaviour analysis
face recognize
flask demo for face recognize
license plate recognize
flask demo for license plate recognize
