I used deepstream to test the FPS. When I used an mp4 as input, the FPS was 25. When I used only one camera as input, the FPS was 25.When I used an mp4 and a camera as input, the mp4’s FPS was still 25, but the camera’s FPS was 0.4.
How can I solve this problem.
The configuration I used:
ubuntu 18.04
gpu :RTX2080Ti —only one gpu
display dirve : 440.59
deepstream_sdk:4.0
cuda:10.1
cudnn:7.6.5
tensorRT : 6.0.1.5
my config_infer_primary_yoloV3.txt:
[property]
gpu-id=0
net-scale-factor=1
#0=RGB, 1=BGR
model-color-format=0
custom-network-config=yolov3.cfg
model-file=yolov3.weights
#model-engine-file=model_b1_int8.engine
labelfile-path=labels.txt
int8-calib-file=yolov3-calibration.table.trt5.1
network-mode=1
num-detected-classes=80
gie-unique-id=1
is-classifier=0
maintain-aspect-ratio=1
parse-bbox-func-name=NvDsInferParseCustomYoloV3
custom-lib-path=nvdsinfer_custom_impl_Yolo/libnvdsinfer_custom_impl_Yolo.so
my deepstream_app_config_yoloV3.txt:
[application]
enable-perf-measurement=1
perf-measurement-interval-sec=5
#gie-kitti-output-dir=streamscl
[tiled-display]
enable=1
rows=1
columns=2
width=1280
height=720
gpu-id=0
#(0): nvbuf-mem-default - Default memory allocated, specific to particular platform
#(1): nvbuf-mem-cuda-pinned - Allocate Pinned/Host cuda memory, applicable for Tesla
#(2): nvbuf-mem-cuda-device - Allocate Device cuda memory, applicable for Tesla
#(3): nvbuf-mem-cuda-unified - Allocate Unified cuda memory, applicable for Tesla
#(4): nvbuf-mem-surface-array - Allocate Surface Array memory, applicable for Jetson
nvbuf-memory-type=0
[source0]
enable=1
#Type - 1=CameraV4L2 2=URI 3=MultiURI
type=3
uri=file://…/…/samples/streams/sample_1080p_h264.mp4
num-sources=1
gpu-id=0
#(0): memtype_device - Memory type Device
#(1): memtype_pinned - Memory type Host Pinned
#(2): memtype_unified - Memory type Unified
cudadec-memtype=0
[source1]
enable=1
#Type - 1=CameraV4L2 2=URI 3=MultiURI
type=3
uri=rtsp://admin:13.168.1.55:554/h264/ch1/main/av_stream
num-sources=1
gpu-id=0
#(0): memtype_device - Memory type Device
#(1): memtype_pinned - Memory type Host Pinned
#(2): memtype_unified - Memory type Unified
cudadec-memtype=0
[sink0]
enable=1
#Type - 1=FakeSink 2=EglSink 3=File
type=2
sync=0
source-id=0
gpu-id=0
nvbuf-memory-type=0
[osd]
enable=1
gpu-id=0
border-width=1
text-size=15
text-color=1;1;1;1;
text-bg-color=0.3;0.3;0.3;1
font=Serif
show-clock=0
clock-x-offset=800
clock-y-offset=820
clock-text-size=12
clock-color=1;0;0;0
nvbuf-memory-type=0
[streammux]
gpu-id=0
##Boolean property to inform muxer that sources are live
live-source=1
batch-size=2
##time out in usec, to wait after the first buffer is available
##to push the batch even if the complete batch is not formed
batched-push-timeout=40000
##Set muxer output width and height
width=1920
height=1080
##Enable to maintain aspect ratio wrt source, and allow black borders, works
##along with width, height properties
enable-padding=0
nvbuf-memory-type=0
#config-file property is mandatory for any gie section.
#Other properties are optional and if set will override the properties set in
#the infer config file.
[primary-gie]
enable=1
gpu-id=0
#model-engine-file=model_b1_int8.engine
labelfile-path=labels.txt
batch-size=2
#Required by the app for OSD, not a plugin property
bbox-border-color0=1;0;0;1
bbox-border-color1=0;1;1;1
bbox-border-color2=0;0;1;1
bbox-border-color3=0;1;0;1
#interval=0
gie-unique-id=1
nvbuf-memory-type=0
config-file=config_infer_primary_yoloV3.txt
[tests]
file-loop=1
The result run :deepstream-app -c deepstream_app_config_yoloV3.txt
local@local-Super-Server:/usr/local/gogo$
local@local-Super-Server:/usr/local/gogo$ deepstream-app -c deepstream_app_config_yoloV3.txt
Creating LL OSD context new
0:00:00.545094945 49044 0x55a8a9ef5d30 INFO nvinfer gstnvinfer.cpp:519:gst_nvinfer_logger:<primary_gie_classifier> NvDsInferContext[UID 1]:initialize(): Trying to create engine from model files
Loading pre-trained weights…
Loading complete!
Total Number of weights read : 61581727
layer inp_size out_size weightPtr
(1) conv-bn-leaky 3 x 320 x 320 32 x 320 x 320 992
(2) conv-bn-leaky 32 x 320 x 320 64 x 160 x 160 19680
(3) conv-bn-leaky 64 x 160 x 160 32 x 160 x 160 21856
(4) conv-bn-leaky 32 x 160 x 160 64 x 160 x 160 40544
(5) skip 64 x 160 x 160 64 x 160 x 160 -
(6) conv-bn-leaky 64 x 160 x 160 128 x 80 x 80 114784
(7) conv-bn-leaky 128 x 80 x 80 64 x 80 x 80 123232
(8) conv-bn-leaky 64 x 80 x 80 128 x 80 x 80 197472
(9) skip 128 x 80 x 80 128 x 80 x 80 -
(10) conv-bn-leaky 128 x 80 x 80 64 x 80 x 80 205920
(11) conv-bn-leaky 64 x 80 x 80 128 x 80 x 80 280160
(12) skip 128 x 80 x 80 128 x 80 x 80 -
(13) conv-bn-leaky 128 x 80 x 80 256 x 40 x 40 576096
(14) conv-bn-leaky 256 x 40 x 40 128 x 40 x 40 609376
(15) conv-bn-leaky 128 x 40 x 40 256 x 40 x 40 905312
(16) skip 256 x 40 x 40 256 x 40 x 40 -
(17) conv-bn-leaky 256 x 40 x 40 128 x 40 x 40 938592
(18) conv-bn-leaky 128 x 40 x 40 256 x 40 x 40 1234528
(19) skip 256 x 40 x 40 256 x 40 x 40 -
(20) conv-bn-leaky 256 x 40 x 40 128 x 40 x 40 1267808
(21) conv-bn-leaky 128 x 40 x 40 256 x 40 x 40 1563744
(22) skip 256 x 40 x 40 256 x 40 x 40 -
(23) conv-bn-leaky 256 x 40 x 40 128 x 40 x 40 1597024
(24) conv-bn-leaky 128 x 40 x 40 256 x 40 x 40 1892960
(25) skip 256 x 40 x 40 256 x 40 x 40 -
(26) conv-bn-leaky 256 x 40 x 40 128 x 40 x 40 1926240
(27) conv-bn-leaky 128 x 40 x 40 256 x 40 x 40 2222176
(28) skip 256 x 40 x 40 256 x 40 x 40 -
(29) conv-bn-leaky 256 x 40 x 40 128 x 40 x 40 2255456
(30) conv-bn-leaky 128 x 40 x 40 256 x 40 x 40 2551392
(31) skip 256 x 40 x 40 256 x 40 x 40 -
(32) conv-bn-leaky 256 x 40 x 40 128 x 40 x 40 2584672
(33) conv-bn-leaky 128 x 40 x 40 256 x 40 x 40 2880608
(34) skip 256 x 40 x 40 256 x 40 x 40 -
(35) conv-bn-leaky 256 x 40 x 40 128 x 40 x 40 2913888
(36) conv-bn-leaky 128 x 40 x 40 256 x 40 x 40 3209824
(37) skip 256 x 40 x 40 256 x 40 x 40 -
(38) conv-bn-leaky 256 x 40 x 40 512 x 20 x 20 4391520
(39) conv-bn-leaky 512 x 20 x 20 256 x 20 x 20 4523616
(40) conv-bn-leaky 256 x 20 x 20 512 x 20 x 20 5705312
(41) skip 512 x 20 x 20 512 x 20 x 20 -
(42) conv-bn-leaky 512 x 20 x 20 256 x 20 x 20 5837408
(43) conv-bn-leaky 256 x 20 x 20 512 x 20 x 20 7019104
(44) skip 512 x 20 x 20 512 x 20 x 20 -
(45) conv-bn-leaky 512 x 20 x 20 256 x 20 x 20 7151200
(46) conv-bn-leaky 256 x 20 x 20 512 x 20 x 20 8332896
(47) skip 512 x 20 x 20 512 x 20 x 20 -
(48) conv-bn-leaky 512 x 20 x 20 256 x 20 x 20 8464992
(49) conv-bn-leaky 256 x 20 x 20 512 x 20 x 20 9646688
(50) skip 512 x 20 x 20 512 x 20 x 20 -
(51) conv-bn-leaky 512 x 20 x 20 256 x 20 x 20 9778784
(52) conv-bn-leaky 256 x 20 x 20 512 x 20 x 20 10960480
(53) skip 512 x 20 x 20 512 x 20 x 20 -
(54) conv-bn-leaky 512 x 20 x 20 256 x 20 x 20 11092576
(55) conv-bn-leaky 256 x 20 x 20 512 x 20 x 20 12274272
(56) skip 512 x 20 x 20 512 x 20 x 20 -
(57) conv-bn-leaky 512 x 20 x 20 256 x 20 x 20 12406368
(58) conv-bn-leaky 256 x 20 x 20 512 x 20 x 20 13588064
(59) skip 512 x 20 x 20 512 x 20 x 20 -
(60) conv-bn-leaky 512 x 20 x 20 256 x 20 x 20 13720160
(61) conv-bn-leaky 256 x 20 x 20 512 x 20 x 20 14901856
(62) skip 512 x 20 x 20 512 x 20 x 20 -
(63) conv-bn-leaky 512 x 20 x 20 1024 x 10 x 10 19624544
(64) conv-bn-leaky 1024 x 10 x 10 512 x 10 x 10 20150880
(65) conv-bn-leaky 512 x 10 x 10 1024 x 10 x 10 24873568
(66) skip 1024 x 10 x 10 1024 x 10 x 10 -
(67) conv-bn-leaky 1024 x 10 x 10 512 x 10 x 10 25399904
(68) conv-bn-leaky 512 x 10 x 10 1024 x 10 x 10 30122592
(69) skip 1024 x 10 x 10 1024 x 10 x 10 -
(70) conv-bn-leaky 1024 x 10 x 10 512 x 10 x 10 30648928
(71) conv-bn-leaky 512 x 10 x 10 1024 x 10 x 10 35371616
(72) skip 1024 x 10 x 10 1024 x 10 x 10 -
(73) conv-bn-leaky 1024 x 10 x 10 512 x 10 x 10 35897952
(74) conv-bn-leaky 512 x 10 x 10 1024 x 10 x 10 40620640
(75) skip 1024 x 10 x 10 1024 x 10 x 10 -
(76) conv-bn-leaky 1024 x 10 x 10 512 x 10 x 10 41146976
(77) conv-bn-leaky 512 x 10 x 10 1024 x 10 x 10 45869664
(78) conv-bn-leaky 1024 x 10 x 10 512 x 10 x 10 46396000
(79) conv-bn-leaky 512 x 10 x 10 1024 x 10 x 10 51118688
(80) conv-bn-leaky 1024 x 10 x 10 512 x 10 x 10 51645024
(81) conv-bn-leaky 512 x 10 x 10 1024 x 10 x 10 56367712
(82) conv-linear 1024 x 10 x 10 21 x 10 x 10 56389237
(83) yolo 21 x 10 x 10 21 x 10 x 10 56389237
(84) route - 512 x 10 x 10 56389237
(85) conv-bn-leaky 512 x 10 x 10 256 x 10 x 10 56521333
(86) upsample 256 x 10 x 10 256 x 20 x 20 -
(87) route - 768 x 20 x 20 56521333
(88) conv-bn-leaky 768 x 20 x 20 256 x 20 x 20 56718965
(89) conv-bn-leaky 256 x 20 x 20 512 x 20 x 20 57900661
(90) conv-bn-leaky 512 x 20 x 20 256 x 20 x 20 58032757
(91) conv-bn-leaky 256 x 20 x 20 512 x 20 x 20 59214453
(92) conv-bn-leaky 512 x 20 x 20 256 x 20 x 20 59346549
(93) conv-bn-leaky 256 x 20 x 20 512 x 20 x 20 60528245
(94) conv-linear 512 x 20 x 20 21 x 20 x 20 60539018
(95) yolo 21 x 20 x 20 21 x 20 x 20 60539018
(96) route - 256 x 20 x 20 60539018
(97) conv-bn-leaky 256 x 20 x 20 128 x 20 x 20 60572298
(98) upsample 128 x 20 x 20 128 x 40 x 40 -
(99) route - 384 x 40 x 40 60572298
(100) conv-bn-leaky 384 x 40 x 40 128 x 40 x 40 60621962
(101) conv-bn-leaky 128 x 40 x 40 256 x 40 x 40 60917898
(102) conv-bn-leaky 256 x 40 x 40 128 x 40 x 40 60951178
(103) conv-bn-leaky 128 x 40 x 40 256 x 40 x 40 61247114
(104) conv-bn-leaky 256 x 40 x 40 128 x 40 x 40 61280394
(105) conv-bn-leaky 128 x 40 x 40 256 x 40 x 40 61576330
(106) conv-linear 256 x 40 x 40 21 x 40 x 40 61581727
(107) yolo 21 x 40 x 40 21 x 40 x 40 61581727
Output blob names :
yolo_83
yolo_95
yolo_107
Total number of layers: 257
Total number of layers on DLA: 0
Building the TensorRT Engine…
0:00:05.155770429 49044 0x55a8a9ef5d30 WARN nvinfer gstnvinfer.cpp:515:gst_nvinfer_logger:<primary_gie_classifier> NvDsInferContext[UID 1]:log(): TensorRT was linked against cuBLAS 10.2.0 but loaded cuBLAS 10.1.0
0:02:00.980585811 49044 0x55a8a9ef5d30 WARN nvinfer gstnvinfer.cpp:515:gst_nvinfer_logger:<primary_gie_classifier> NvDsInferContext[UID 1]:log(): TensorRT was linked against cuBLAS 10.2.0 but loaded cuBLAS 10.1.0
Building complete!
0:02:02.974513606 49044 0x55a8a9ef5d30 INFO nvinfer gstnvinfer.cpp:519:gst_nvinfer_logger:<primary_gie_classifier> NvDsInferContext[UID 1]:generateTRTModel(): Storing the serialized cuda engine to file at /usr/local/gogo/model_b2_int8.engine
Deserialize yoloLayerV3 plugin: yolo_83
Deserialize yoloLayerV3 plugin: yolo_95
Deserialize yoloLayerV3 plugin: yolo_107
0:02:03.988662431 49044 0x55a8a9ef5d30 WARN nvinfer gstnvinfer.cpp:515:gst_nvinfer_logger:<primary_gie_classifier> NvDsInferContext[UID 1]:log(): TensorRT was linked against cuBLAS 10.2.0 but loaded cuBLAS 10.1.0
0:02:03.992511034 49044 0x55a8a9ef5d30 WARN nvinfer gstnvinfer.cpp:515:gst_nvinfer_logger:<primary_gie_classifier> NvDsInferContext[UID 1]:log(): TensorRT was linked against cuBLAS 10.2.0 but loaded cuBLAS 10.1.0
cb_sourcesetup set 100 latency
Runtime commands:
h: Print this help
q: Quit
p: Pause
r: Resume
NOTE: To expand a source in the 2D tiled display and view object details, left-click on the source.
To go back to the tiled display, right-click anywhere on the window.
**PERF: FPS 0 (Avg) FPS 1 (Avg)
**PERF: 0.00 (0.00) 0.00 (0.00)
** INFO: <bus_callback:189>: Pipeline ready
Creating LL OSD context new
** INFO: <bus_callback:175>: Pipeline running
**PERF: 25.74 (25.74) 26.34 (26.34)
**PERF: 0.00 (25.74) 24.87 (25.58)
**PERF: 0.09 (1.01) 25.02 (25.39)
**PERF: 0.17 (0.73) 25.01 (25.29)
**PERF: 0.17 (0.60) 25.01 (25.24)
**PERF: 0.26 (0.55) 25.01 (25.20)
**PERF: 0.18 (0.49) 25.02 (25.17)
**PERF: 0.27 (0.47) 25.01 (25.15)
**PERF: 0.00 (0.47) 24.87 (25.12)
q
Quitting
App run successful
I hope to provide some useful advice,thanks.