search

inayatkh / realTimeObjectDetection

This is a demonstration of a MobileNetSDD object detector for achieving a very high frame per secons(FPS)
4 stars 1 forks source link

some issue，could you help me？ #2

Open enyblock opened 6 years ago

enyblock commented 6 years ago

hi， thanks for you gread work.

now i used this project on my pi3, i already installed caffe-cup in my pi3,then i changed the code (ssdcaffe/ssd.py line 53) to ´self.USEGPU = False´ then i run the command, i got the error, could you help me ? thanks a lot.

(cv) pi@raspberrypi:~/Downloads/realTimeObjectDetection-master $ python deepDetectVideo.py -v ./videos//Kids-Truck-Bus-video.mkv [INFO] Using CPU Only - 2017-12-02 01:28:17.190900 [INFO] Loading caffe model: ./deepModel/MobileNetSSD_deploy.caffemodel - 2017-12-02 01:28:17.191151 WARNING: Logging before InitGoogleLogging() is written to STDERR W1202 01:28:17.191323 3881 _caffe.cpp:135] DEPRECATION WARNING - deprecated use of Python interface W1202 01:28:17.191468 3881 _caffe.cpp:136] Use this instead (with the named "weights" parameter): W1202 01:28:17.191519 3881 _caffe.cpp:138] Net('./deepModel/MobileNetSSD_deploy.prototxt.txt', 1, weights='./deepModel/MobileNetSSD_deploy.caffemodel') [libprotobuf ERROR google/protobuf/text_format.cc:298] Error parsing text-format caffe.NetParameter: 1177:17: Message type "caffe.LayerParameter" has no field named "permute_param". F1202 01:28:17.203552 3881 upgrade_proto.cpp:88] Check failed: ReadProtoFromTextFile(param_file, param) Failed to parse NetParameter file: ./deepModel/MobileNetSSD_deploy.prototxt.txt Check failure stack trace: Aborted

inayatkh commented 6 years ago

Hi Thanks for using my code. Basically you have to use the following version of Caffe , its import to check out the ssd of its version

https://github.com/inayatkh/caffe/tree/ssd

I hope this will solve your problem

Inayat

enyblock commented 6 years ago

ok, i will try it.

btw, did you run the code on pi ? 14fps ?