vijendra1125 / custom-faster-rcnn-using-tensorfow-object-detection-api Goto Github PK

working with showing results on which contains only single damage area labled in image. and multiple damage area labeled in image is not showing results with multiple bounding box why?

@vijendra1125 @TannerGilbert

i get error in adding path and string_int_label_map_pb2.. i already install protobuff.. can you please tell me more elabaorately.. I am using windows.. help will be grateful

c_api.TF_GetCode(self.status.status))
tensorflow.python.framework.errors_impl.NotFoundError: /media/aman/48aca5a1-5813-4608-90a1-8660bc24c17e/Tensorflow_Object_detection_API-Custom_Faster_RCNN/dataset/VOC2007/ImageSets/Main/aeroplane_train.txt; No such file or directory

Hi,
Currently myself working in Object Detection , I have used SSD config for Tensor-flow(TF) for training and testing sample video it takes 5fps for one second. Now i want to use the Faster RCNN in desktop , but noticing few performance issue which Executes only 3 Images per second, whereas its slower than SSD config .

Can expect a better performance and accuracy in Faster RCNN when compared to SSD TF?
Is there any available for Lite version of Faster RCNN?

On line 157, in create_pascal_tf_record.py (from extra folder), when I run xml_str = fid.read(), the resultant xml_str is empty. And hence, when I run xml = etree.fromstring(xml_str), it's throwing the following error:

    xml = etree.fromstring(xml_str)
  File "src/lxml/etree.pyx", line 3237, in lxml.etree.fromstring
  File "src/lxml/parser.pxi", line 1896, in lxml.etree._parseMemoryDocument
  File "src/lxml/parser.pxi", line 1777, in lxml.etree._parseDoc
  File "src/lxml/parser.pxi", line 1082, in lxml.etree._BaseParser._parseUnicodeDoc
  File "src/lxml/parser.pxi", line 615, in lxml.etree._ParserContext._handleParseResultDoc
  File "src/lxml/parser.pxi", line 725, in lxml.etree._handleParseResult
  File "src/lxml/parser.pxi", line 654, in lxml.etree._raiseParseError
  File "<string>", line 1
lxml.etree.XMLSyntaxError: Document is empty, line 1, column 1

I am using the VOC 2012 data and I check the first image and its xml in labelimg and it's not empty.
So why am I getting this error?