SK-TDA4VM: EdgeAI Apps don't display DL result when multi-inference using 1080P input

Taekmin Lee

Part Number: SK-TDA4VM
Other Parts Discussed in Thread: TDA4VM

Hi,

I tested EdgeAI Apps using 'single_input_multi_inference' config.
It works using single 720P video input → 4ch inference → mosaic display. but I changed input video 720P to 1080P, DL result don't display.
The app's log shows DL inference is working. like below:

[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 14.00 ms (avg 12.24 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 84.67 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 13.02 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 30.00 ms (avg 11.31 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 87.57 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 12.96 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 20.00 ms (avg 10.70 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 84.78 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 12.81 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  8.00 ms (avg 11.22 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 86.39 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 13.22 fps

And I tested '1080P video input → 1/2/3ch inference → mosaic display' and it works. so there is no problem in test 1080P video file.

Also I tested RTSP input instead of video file, but the result is same of using video file. 1080P input only work 3ch inference.
I know that TDA4VM has 'D5520MP2' decoder. It support 4x1080p60 and 8x1080p30 decoder. so I think it's not hardware problem.

How can i fix this problem?

Regards,
Lee.

over 3 years ago

0 Shyam Jagannathan over 3 years ago

TI__Genius 10355 points

Hi Lee,

Can you please provide some more details,

1. Which version of EdgeAI SDK are you using?

2. Are you running Python / C++ app?

3. Can you please run the app with -n option if its Python and -n -l 0 if its C++ app?

4. Also please run the same from UART console (not SSH terminal) and provide us the log.

Regards,
Shyam

0 Taekmin Lee over 3 years ago in reply to Shyam Jagannathan

Expert 1445 points

Hi Shyam,

1. Processor SDK Linux for Edge AI 08.00.01

2. I used C++ based app. Also i tested python based apps, the result is same.

3. I executed the app with -n -l options.

4. I used UART consol.

I attached the c++ based app's log file.

Fullscreen 4ch_multi_infer.log Download

 _____                    _____           _         _
|  _  |___ ___ ___ ___   |  _  |___ ___  |_|___ ___| |_
|     |  _| .'| . | . |  |   __|  _| . | | | -_|  _|  _|
|__|__|_| |__,|_  |___|  |__|  |_| |___|_| |___|___|_|
              |___|                    |___|

Arago Project http://arago-project.org j7-evm ttyS2

Arago 2021.09 j7-evm ttyS2

j7-evm login: root
Last login: Tue Apr 12 01:57:54 UTC 2022

[   30.106013] TI DP83867 46000f00.mdio:00: Downshift occurred from negotiated speed 1Gbps to actual speed 100Mbps, check cabling!
[   30.117507] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
[   30.123792] am65-cpsw-nuss 46000000.ethernet eth0: Link is Up - 100Mbps/Full (downshifted) - flow control off

[   37.267355]  sda:
root@j7-evm:/opt/edge_ai_apps#
root@j7-evm:/opt/edge_ai_apps#
root@j7-evm:/opt/edge_ai_apps# [   44.195723] Initializing XFRM netlink socket
[   46.975027] process 'docker/tmp/qemu-check996738659/check' started with executable stack

root@j7-evm:/opt/edge_ai_apps# cd apps_cpp/
root@j7-evm:/opt/edge_ai_apps/apps_cpp# ./bin/Release/app_edgeai -n -v ../configs/object_detection.yaml

 Number of subgraphs:1 , 126 nodes delegated out of 126 nodes

APP: Init ... !!!
MEM: Init ... !!!
MEM: Initialized DMA HEAP (fd=5) !!!
MEM: Init ... Done !!!
IPC: Init ... !!!
IPC: Init ... Done !!!
REMOTE_SERVICE: Init ... !!!
REMOTE_SERVICE: Init ... Done !!!
    73.642397 s: GTC Frequency = 200 MHz
APP: Init ... Done !!!
    73.648981 s:  VX_ZONE_INIT:Enabled
    73.649007 s:  VX_ZONE_ERROR:Enabled
    73.649017 s:  VX_ZONE_WARNING:Enabled
    73.651578 s:  VX_ZONE_INIT:[tivxInitLocal:130] Initialization Done !!!
    73.653580 s:  VX_ZONE_INIT:[tivxHostInitLocal:86] Initialization Done for HOST !!!

 Number of subgraphs:1 , 126 nodes delegated out of 126 nodes


 Number of subgraphs:1 , 126 nodes delegated out of 126 nodes


 Number of subgraphs:1 , 126 nodes delegated out of 126 nodes

[   67.990161] unclassified data detected!
[   68.099640] unclassified data detected!
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  8.00 ms (avg  8.00 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 15.00 ms (avg 15.00 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 20.00 ms (avg 20.00 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 12.00 ms (avg 12.00 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  8.00 ms (avg 10.00 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 21.00 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 47.62 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 14.00 ms (avg 11.00 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 44.00 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 22.73 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  8.00 ms (avg  9.33 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 26.50 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 39.43 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 11.00 ms (avg 13.00 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 81.00 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 12.35 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  8.00 ms (avg 14.00 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 98.00 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 10.20 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  8.75 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 32.67 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 33.70 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  9.67 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 58.00 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 18.31 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 11.00 ms (avg 12.33 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 88.00 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 11.44 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 13.00 ms (avg  9.60 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 11.00 ms (avg 10.00 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 67.33 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 16.08 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 13.00 ms (avg 13.67 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 105.00 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate':  9.57 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 12.00 ms (avg 12.25 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  9.40 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg 12.00 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 107.00 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate':  9.38 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  6.00 ms (avg 10.80 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 89.00 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 14.18 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 120.00 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate':  9.44 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 93.75 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 13.51 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 13.00 ms (avg 12.40 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 94.75 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 20.23 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  9.00 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 77.20 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 28.99 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  9.00 ms (avg  9.00 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  8.00 ms (avg 10.33 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 95.80 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 12.97 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 119.00 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 25.93 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 81.67 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 25.76 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  8.00 ms (avg  9.33 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 21.00 ms (avg 13.83 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 101.20 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 17.76 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 16.00 ms (avg  9.88 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 100.20 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 28.75 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  9.00 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 85.50 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 37.85 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  8.00 ms (avg 13.00 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  8.75 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 92.50 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 18.20 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 77.86 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 36.90 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 80.14 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 24.09 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  9.56 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 73.88 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 25.25 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 10.00 ms (avg 10.29 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 103.17 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 12.00 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  9.00 ms (avg 12.50 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 88.86 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 17.74 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 12.00 ms (avg  9.80 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 73.44 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 24.03 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  8.56 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 81.88 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 33.43 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  8.00 ms (avg 12.00 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 86.88 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 17.23 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  8.00 ms (avg 10.00 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 100.29 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 12.00 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  9.55 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 72.70 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 23.14 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 10.00 ms (avg  8.70 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 16.00 ms (avg 12.40 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 87.00 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 16.58 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 85.89 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 30.65 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  9.33 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 72.00 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 22.44 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  9.67 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 102.25 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 11.58 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 12.00 ms (avg 12.36 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 86.20 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 16.19 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 13.00 ms (avg  9.09 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 86.00 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 28.74 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  9.15 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 74.33 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 21.40 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  9.40 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 100.22 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 11.62 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 10.00 ms (avg 12.17 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 85.00 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 15.96 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  8.92 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 85.73 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 27.22 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  8.00 ms (avg  9.07 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 74.77 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 20.71 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  8.00 ms (avg  9.27 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 97.90 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 11.75 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  6.00 ms (avg 11.69 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 85.08 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 15.60 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  8.77 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 85.25 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 25.99 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  9.08 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 95.73 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 11.91 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  8.00 ms (avg  9.00 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 77.14 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 19.90 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg 11.36 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 83.46 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 15.60 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  8.64 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 85.08 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 24.92 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 12.00 ms (avg  9.19 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 77.53 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 19.37 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  9.00 ms (avg  9.08 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 97.33 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 11.65 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  9.00 ms (avg 11.20 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 83.50 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 15.34 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  8.53 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 83.64 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 24.24 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 11.00 ms (avg  9.21 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 12.00 ms (avg  9.35 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 97.38 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 11.54 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 79.25 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 18.76 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 12.00 ms (avg 11.25 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 84.80 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 14.96 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 12.00 ms (avg  8.75 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 84.40 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 23.33 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  9.07 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 94.21 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 12.06 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  9.22 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 79.35 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 18.38 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  8.00 ms (avg 11.06 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 84.94 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 14.75 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  9.00 ms (avg  8.76 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 84.38 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 22.61 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  8.94 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 95.33 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 11.86 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  8.00 ms (avg 10.89 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 84.71 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 14.60 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 12.00 ms (avg  8.94 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 84.41 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 21.97 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 13.00 ms (avg  9.42 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 80.61 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 17.90 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg 10.68 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 82.33 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 15.12 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 11.00 ms (avg  9.06 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 96.06 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 11.70 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  9.00 ms (avg  9.40 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 81.05 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 17.55 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 12.00 ms (avg  9.11 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 85.50 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 21.29 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg 10.50 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 81.74 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 15.06 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  8.94 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 95.35 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 11.71 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  8.00 ms (avg  9.33 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 81.40 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 17.24 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  9.00 ms (avg  9.10 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 85.37 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 20.80 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 10.00 ms (avg 10.48 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 84.55 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 14.67 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  8.00 ms (avg  8.89 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 94.22 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 11.80 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  8.00 ms (avg  9.27 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 80.95 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 17.08 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  9.00 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 85.60 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 20.32 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 12.00 ms (avg 10.55 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  9.00 ms (avg  8.90 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  9.00 ms (avg  9.26 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 81.05 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 16.85 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 93.95 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 11.77 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 85.29 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 14.45 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  8.91 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 86.05 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 19.85 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  9.00 ms (avg 10.48 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference': 12.00 ms (avg  9.05 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 93.50 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 11.77 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  8.00 ms (avg  9.21 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 81.48 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 16.60 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 85.36 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 14.31 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  7.00 ms (avg  8.83 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 86.14 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 19.46 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  9.00 ms (avg 10.42 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 84.17 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 14.44 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  8.00 ms (avg  9.16 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 80.96 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 16.51 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  8.00 ms (avg  8.79 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Time for 'dl-inference':  9.00 ms (avg  9.05 ms)
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 85.04 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 19.33 fps
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'total time': 93.71 ms
[UTILS] [Model Name:   TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320] Metric 'framerate': 11.70 fps

Regards,
Lee.

0 Shyam Jagannathan over 3 years ago in reply to Taekmin Lee

TI__Genius 10355 points

Hi Lee,

Thanks for the information and logs. I have copied an expert who will help with your issue. Internally we will try and replicate this as well.

Regards,
Shyam

0 Rahul Ravikumar over 3 years ago in reply to Shyam Jagannathan

TI__Expert 4218 points

Hi Lee,

Can you set width as 1088 in the config file and try
I observed that h264parse is aligning the 1080p buffers to 1920x1088

Regards
Rahul T R

0 Taekmin Lee over 3 years ago in reply to Rahul Ravikumar

Expert 1445 points

Hi Rahul,

Did you mean output's width?
I set output's width 1088 in config file, result is same. nothing to change.

I attached file that used in my test(config.yaml, sample_fhd.mp4)
Please check using this files.

fhd_multi_test.zip.

Regards,
Lee.

0 Rahul Ravikumar over 3 years ago in reply to Taekmin Lee

TI__Expert 4218 points

Hi Lee,

I meant the input width
This will be used for some calculations
please change input width to 1088

Regards
Rahul T R

0 Taekmin Lee over 3 years ago in reply to Rahul Ravikumar

Expert 1445 points

Hi Rahul,

I modified input width 1080 to 1088, result is same. DL-inference result doesn't display.

Can you test it with the file attached to the above article?

title: "Single Input, Multi Inference Demo"
log_level: 2
inputs:
    input0:
        source: /opt/edge_ai_apps/data/videos/video_0000_h264.mp4
        format: h264
        width: 1280
        height: 720
        framerate: 25
        loop: False
    input1:
        source: /opt/edge_ai_apps/data/videos/fhd_video_0000_h264.mp4
        format: h264
        width: 1920
        height: 1088
        framerate: 25
        loop: False
models:
    model0:
        model_path: /opt/model_zoo/TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320
        viz_threshold: 0.6
    model1:
        model_path: /opt/model_zoo/TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320
        viz_threshold: 0.6
    model2:
        model_path: /opt/model_zoo/TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320
        viz_threshold: 0.6
    model3:
        model_path: /opt/model_zoo/TFL-OD-2020-ssdLite-mobDet-DSP-coco-320x320
        viz_threshold: 0.6
outputs:
    output0:
        sink: kmssink
        width: 1920
        height: 1080

flows:
    flow0:
        input: input1
        models: [model0, model1, model2, model3]
        outputs: [output0, output0, output0, output0]
        mosaic:
            mosaic0:
                width:  960
                height: 540
                pos_x:  0
                pos_y:  0
            mosaic1:
                width:  960
                height: 540
                pos_x:  960
                pos_y:  0
            mosaic2:
                width:  960
                height: 540
                pos_x:  0
                pos_y:  540
            mosaic3:
                width:  960
                height: 540
                pos_x:  960
                pos_y:  540

Regards,
Lee.

0 Rahul Ravikumar over 3 years ago in reply to Taekmin Lee

TI__Expert 4218 points

Hi Lee,

Are you trying python apps or cpp apps
Can you try with python apps with -n option
and share the gstreamer pipeline printed on the console

if you are using minicom, add -w option to avoid line wrapping
I want investigate the pipeline constructed

Regards
Rahul T R

0 Taekmin Lee over 3 years ago in reply to Rahul Ravikumar

Expert 1445 points

Hi Rahul,

I tried cpp apps. here is python app's log.

root
Last login: Thu Apr 14 05:55:57 UTC 2022
root@j7-evm:/opt/edge_ai_apps# cd apps_python/
root@j7-evm:/opt/edge_ai_apps/apps_python# ./app_edgeai.py -n ../configs/single_input_multi_infer_fhd.yaml

 Number of subgraphs:1 , 126 nodes delegated out of 126 nodes

APP: Init ... !!!
MEM: Init ... !!!
MEM: Initialized DMA HEAP (fd=5) !!!
MEM: Init ... Done !!!
IPC: Init ... !!!
IPC: Init ... Done !!!
REMOTE_SERVICE: Init ... !!!
REMOTE_SERVICE: Init ... Done !!!
   689.446870 s: GTC Frequency = 200 MHz
APP: Init ... Done !!!
   689.452185 s:  VX_ZONE_INIT:Enabled
   689.452208 s:  VX_ZONE_ERROR:Enabled
   689.452214 s:  VX_ZONE_WARNING:Enabled
   689.454729 s:  VX_ZONE_INIT:[tivxInitLocal:130] Initialization Done !!!
   689.456730 s:  VX_ZONE_INIT:[tivxHostInitLocal:86] Initialization Done for HOST !!!

 Number of subgraphs:1 , 126 nodes delegated out of 126 nodes


 Number of subgraphs:1 , 126 nodes delegated out of 126 nodes


 Number of subgraphs:1 , 126 nodes delegated out of 126 nodes

[GST SRC STR]
[FLOW 0]
filesrc location=/opt/edge_ai_apps/data/videos/fhd_video_0000_h264.mp4 ! qtdemux ! h264parse ! v4l2h264dec ! video/x-raw, format=NV12  ! tee name=tee_split0
tee_split0. ! queue ! tiovxmultiscaler name=split_01
tee_split0. ! queue ! tiovxmultiscaler name=split_02
tee_split0. ! queue ! tiovxmultiscaler name=split_03
tee_split0. ! queue ! tiovxmultiscaler name=split_04
split_01. ! queue ! video/x-raw, width=480, height=320 ! tiovxmultiscaler ! video/x-raw, width=320, height=320 ! tiovxdlpreproc data-type=10 channel-order=1 mean-0=128.000000 mean-1=128.000000 mean-2=128.000000 scale-0=0.007812 scale-1=0.007812 scale-2=0.007812 tensor-format=rgb out-pool-size=4 ! application/x-tensor-tiovx ! appsink name=pre_0 max-buffers=2 drop=true
split_01. ! queue ! video/x-raw, width=960, height=540 ! tiovxcolorconvert target=1 out-pool-size=4 ! video/x-raw, format=RGB ! appsink name=sen_0 max-buffers=2 drop=true
split_02. ! queue ! video/x-raw, width=480, height=320 ! tiovxmultiscaler ! video/x-raw, width=320, height=320 ! tiovxdlpreproc data-type=10 channel-order=1 mean-0=128.000000 mean-1=128.000000 mean-2=128.000000 scale-0=0.007812 scale-1=0.007812 scale-2=0.007812 tensor-format=rgb out-pool-size=4 ! application/x-tensor-tiovx ! appsink name=pre_1 max-buffers=2 drop=true
split_02. ! queue ! video/x-raw, width=960[  684.247694] unclassified data detected!
, height=540 ! tiovxcolorconvert target=1 out-pool-size=4 ! video/x-raw, format=RGB ! appsink name=sen_1 max-buffers=2 drop=true
split_03. ! queue ! video/x-raw, width=480, height=320 ! tiovxmultiscaler ! video/x-raw, width=320, height=320 ! tiovxdlpreproc data-type=10 channel-order=1 mean-0=128.000000 mean-1=128.000000 mean-2=128.000000 scale-0=0.007812 scale-1=0.007812 scale-2=0.007812 tensor-format=rgb out-pool-size=4 ! application/x-tensor-tiovx ! appsink name=pre_2 max-buffers=2 drop=true
split_03. ! queue ! video/x-raw, width=960, height=540 ! tiovxcolorconvert target=1 out-pool-size=4 ! video/x-raw, format=RGB ! appsink name=sen_2 max-buffers=2 drop=true
split_04. ! queue ! video/x-raw, width=480, height=320 ! tiovxmultiscaler ! video/x-raw, wid[  684.320221] unclassified data detected!
th=320, height=320 ! tiovxdlpreproc data-type=10 channel-order=1 mean-0=128.000000 mean-1=128.000000 mean-2=128.000000 scale-0=0.007812 scale-1=0.007812 scale-2=0.007812 tensor-format=rgb out-pool-size=4 ! application/x-tensor-tiovx ! appsink name=pre_3 max-buffers=2 drop=true
split_04. ! queue ! video/x-raw, width=960, height=540 ! tiovxcolorconvert target=1 out-pool-size=4 ! video/x-raw, format=RGB ! appsink name=sen_3 max-buffers=2 drop=true

[GST SINK STR]
appsrc format=GST_FORMAT_TIME is-live=true block=true do-timestamp=true name=post_0 ! tiovxcolorconvert ! video/x-raw,format=NV12, width=960, height=540 ! queue ! mosaic_0.sink_0
appsrc format=GST_FORMAT_TIME is-live=true block=true do-timestamp=true name=post_1 ! tiovxcolorconvert ! video/x-raw,format=NV12, width=960, height=540 ! queue ! mosaic_0.sink_1
appsrc format=GST_FORMAT_TIME is-live=true block=true do-timestamp=true name=post_2 ! tiovxcolorconvert ! video/x-raw,format=NV12, width=960, height=540 ! queue ! mosaic_0.sink_2
appsrc format=GST_FORMAT_TIME is-live=true block=true do-timestamp=true name=post_3 ! tiovxcolorconvert ! video/x-raw,format=NV12, width=960, height=540 ! queue ! mosaic_0.sink_3
appsrc format=GST_FORMAT_TIME block=true num-buffers=1 name=background_0 ! tiovxcolorconvert ! video/x-raw,format=NV12, width=1920, height=1080 ! queue ! mosaic_0.background
tiovxmosaic name=mosaic_0
sink_0::startx=0  sink_0::starty=0  sink_0::width=960   sink_0::height=540
sink_1::startx=960  sink_1::starty=0  sink_1::width=960   sink_1::height=540
sink_2::startx=0  sink_2::starty=540  sink_2::width=960   sink_2::height=540
sink_3::startx=960  sink_3::starty=540  sink_3::width=960   sink_3::height=540
! video/x-raw,format=NV12, width=1920, height=1080 ! kmssink sync=false driver-name=tidss

Regards,
Lee.

0 Rahul Ravikumar over 3 years ago in reply to Taekmin Lee

TI__Expert 4218 points

Hi Lee,

Your config file is working fine at my end
can you please share how you scaled the video to 1920x1080

I used below ffmpeg command:

ffmpeg -i video_0001_h264.mp4 -c:v libx264 -bf 0 -vb 10000k -qmax 24 -vf fps=fps=30 -vf scale=1920:1080 -an video_0001_h264_b.mp4

can you try with this?

Regards
Rahul T R

0 Taekmin Lee over 3 years ago in reply to Rahul Ravikumar

Expert 1445 points

Hi Rahul,

I scaled the video HD to FHD using a regular encoder program. The FHD video file work well without 4ch.
As mentioned, the attachment also includes a video file.

hum... Seeing that FHD 4Ch works normally in your environment, I think there is a problem with my environment.
I'll make new SD card and check it again.

Regards,
Lee.

0 Rahul Ravikumar over 3 years ago in reply to Taekmin Lee

TI__Expert 4218 points

Hi Lee,

Which SDK version you are using
8.1?

My environment might be different
than 8.1, I will give a try with 8.1

Regards
Rahul T R

0 Taekmin Lee over 3 years ago in reply to Rahul Ravikumar

Expert 1445 points

Hi Rahul,

I use SDK v08.01.00.02.

Regards,
Lee.

0 Taekmin Lee over 3 years ago in reply to Taekmin Lee

Expert 1445 points

Hi, update this thread.

The SDK 8.2 was released.
So, I tested 4ch inference with same setting on board that installed SDK 8.2. It works well.

(1 FHD → 4ch Inference → Mosaic)

I think it's a problem with some of the elements of SDK 8.1.

Regards,
Lee.

Processors

Processors forum

SK-TDA4VM: EdgeAI Apps don't display DL result when multi-inference using 1080P input