PyPostYolo.PyPostYolo Class Reference

Simple YOLO DNN post-processor written in python. More...

Public Member Functions
def	__init__ (self)
	[Optional] Constructor More...
def	init (self)
	[Optional] JeVois parameters initialization More...
def	freeze (self, doit)
	[Optional] Freeze some parameters that should not be changed at runtime. More...
def	loadClasses (self, filename)
	[Optional] Parameter callback: Load class names when 'classes' parameter value is changed by model zoo More...
def	setDetectType (self, dt)
	[Optional] Parameter callback: set type of object detector More...
def	process (self, outs, preproc)
	[Required] Main processing function: parse network output blobs and store resulting labels and scores locally. More...
def	getLabel (self, id, conf)
	Helper to get class name and confidence as a clean string, and a color that varies with class name. More...
def	report (self, outimg, helper, overlay, idle)
	[Optional] Report the latest results obtained by process() by drawing them outimg is None or a RawImage to draw into when in Legacy mode (drawing to an image sent to USB) helper is None or a GUIhelper to do OpenGL drawings when in JeVois-Pro GUI mode overlay is True if user wishes to see overlay text idle is true if keyboard/mouse have been idle for a while, which typically would reduce what is displayed More...

Public Attributes
	classIds
	confidences
	boxes
	classmap
	yolopp
	classoffset
	classes
	detecttype
	nms
	maxnbox
	cthresh
	dthresh
	imageh

Detailed Description

Simple YOLO DNN post-processor written in python.

Compare this code to the C++ PostProcessorDetect (which has more functionality than here):

• Abstract base: https://github.com/jevois/jevois/blob/master/include/jevois/DNN/PostProcessor.H
• Header: https://github.com/jevois/jevois/blob/master/include/jevois/DNN/PostProcessorDetect.H
• Implementation: https://github.com/jevois/jevois/blob/master/src/jevois/DNN/PostProcessorDetect.C

Instead of re-inventing the wheel, this code uses the YOLO post-processor that we have implemented in C++, as that C++ code is quite complex and multi-threaded for speed.

Author

Laurent Itti

Email:

itti@usc.edu

Address:

University of Southern California, HNB-07A, 3641 Watt Way, Los Angeles, CA 90089-2520, USA

Copyright

Copyright (C) 2022 by Laurent Itti, iLab and the University of Southern California

Main URL:

http://jevois.org

Support URL:

http://jevois.org/doc

Other URL:

http://iLab.usc.edu

License:

GPL v3

Distribution:

Unrestricted

Restrictions:

None

Definition at line 30 of file PyPostYolo.py.

Constructor & Destructor Documentation

__init__()

def PyPostYolo.PyPostYolo.__init__

(

self

)

[Optional] Constructor

Definition at line 33 of file PyPostYolo.py.

Member Function Documentation

freeze()

def PyPostYolo.PyPostYolo.freeze	(		self,
			doit
	)

[Optional] Freeze some parameters that should not be changed at runtime.

The JeVois core will call this with doit being either True or False

Definition at line 91 of file PyPostYolo.py.

References PyPostClassify.PyPostClassify.classes, PyPostDAMOyolo.PyPostDAMOyolo.classes, PyPostYolo.PyPostYolo.classes, Darknet.classes, Yolo.classes, PyPostYolo.PyPostYolo.detecttype, and PyPostYolo.PyPostYolo.yolopp.

getLabel()

def PyPostYolo.PyPostYolo.getLabel	(		self,
			id,
			conf
	)

Helper to get class name and confidence as a clean string, and a color that varies with class name.

Definition at line 170 of file PyPostYolo.py.

References PyPostClassify.PyPostClassify.classmap, PyPostDAMOyolo.PyPostDAMOyolo.classmap, and PyPostYolo.PyPostYolo.classmap.

Referenced by PyPostYolo.PyPostYolo.report().

init()

def PyPostYolo.PyPostYolo.init

(

self

)

[Optional] JeVois parameters initialization

Definition at line 49 of file PyPostYolo.py.

loadClasses()

def PyPostYolo.PyPostYolo.loadClasses	(		self,
			filename
	)

[Optional] Parameter callback: Load class names when 'classes' parameter value is changed by model zoo

Definition at line 98 of file PyPostYolo.py.

References PyPostClassify.PyPostClassify.classmap, PyPostDAMOyolo.PyPostDAMOyolo.classmap, PyPostYolo.PyPostYolo.classmap, and split().

process()

def PyPostYolo.PyPostYolo.process	(		self,
			outs,
			preproc
	)

[Required] Main processing function: parse network output blobs and store resulting labels and scores locally.

outs is a list of numpy arrays for the network's outputs. preproc is a handle to the pre-processor that was used, useful to recover transforms from original image to cropped/resized network inputs.

Definition at line 115 of file PyPostYolo.py.

References PyPostDAMOyolo.PyPostDAMOyolo.boxes, PyPostYolo.PyPostYolo.boxes, PyPostDAMOyolo.PyPostDAMOyolo.classIds, PyPostYolo.PyPostYolo.classIds, PyPostDAMOyolo.PyPostDAMOyolo.confidences, and PyPostYolo.PyPostYolo.confidences.

report()

def PyPostYolo.PyPostYolo.report	(		self,
			outimg,
			helper,
			overlay,
			idle
	)

[Optional] Report the latest results obtained by process() by drawing them outimg is None or a RawImage to draw into when in Legacy mode (drawing to an image sent to USB) helper is None or a GUIhelper to do OpenGL drawings when in JeVois-Pro GUI mode overlay is True if user wishes to see overlay text idle is true if keyboard/mouse have been idle for a while, which typically would reduce what is displayed

Note that report() is called on every frame even though the network may run slower or take some time to load and initialize, thus you should be prepared for report() being called even before process() has ever been called (i.e., create some class member variables to hold the reported results, initialize them to some defaults in your constructor, report their current values here, and update their values in process()).

Definition at line 189 of file PyPostYolo.py.

References PyPostDAMOyolo.PyPostDAMOyolo.boxes, PyPostYolo.PyPostYolo.boxes, PyPostDAMOyolo.PyPostDAMOyolo.classIds, PyPostYolo.PyPostYolo.classIds, PyPostDAMOyolo.PyPostDAMOyolo.confidences, PyPostYolo.PyPostYolo.confidences, PyPostDAMOyolo.PyPostDAMOyolo.getLabel(), and PyPostYolo.PyPostYolo.getLabel().

setDetectType()

def PyPostYolo.PyPostYolo.setDetectType	(		self,
			dt
	)

[Optional] Parameter callback: set type of object detector

Definition at line 106 of file PyPostYolo.py.

Member Data Documentation

boxes

PyPostYolo.PyPostYolo.boxes

Definition at line 37 of file PyPostYolo.py.

Referenced by PyPostYolo.PyPostYolo.process(), and PyPostYolo.PyPostYolo.report().

classes

PyPostYolo.PyPostYolo.classes

Definition at line 56 of file PyPostYolo.py.

Referenced by PyPostYolo.PyPostYolo.freeze(), PyDetectionDNN.PyDetectionDNN.postprocess(), and PyClassificationDNN.PyClassificationDNN.process().

classIds

PyPostYolo.PyPostYolo.classIds

Definition at line 35 of file PyPostYolo.py.

Referenced by PyPostYolo.PyPostYolo.process(), and PyPostYolo.PyPostYolo.report().

classmap

PyPostYolo.PyPostYolo.classmap

Definition at line 40 of file PyPostYolo.py.

Referenced by PyPostYolo.PyPostYolo.getLabel(), and PyPostYolo.PyPostYolo.loadClasses().

classoffset

PyPostYolo.PyPostYolo.classoffset

Definition at line 52 of file PyPostYolo.py.

confidences

PyPostYolo.PyPostYolo.confidences

Definition at line 36 of file PyPostYolo.py.

Referenced by PyPostYolo.PyPostYolo.process(), and PyPostYolo.PyPostYolo.report().

cthresh

PyPostYolo.PyPostYolo.cthresh

Definition at line 75 of file PyPostYolo.py.

detecttype

PyPostYolo.PyPostYolo.detecttype

Definition at line 61 of file PyPostYolo.py.

Referenced by PyPostYolo.PyPostYolo.freeze().

dthresh

PyPostYolo.PyPostYolo.dthresh

Definition at line 79 of file PyPostYolo.py.

imageh

PyPostYolo.PyPostYolo.imageh

Definition at line 124 of file PyPostYolo.py.

maxnbox

PyPostYolo.PyPostYolo.maxnbox

Definition at line 70 of file PyPostYolo.py.

nms

PyPostYolo.PyPostYolo.nms

Definition at line 66 of file PyPostYolo.py.

yolopp

PyPostYolo.PyPostYolo.yolopp

Definition at line 45 of file PyPostYolo.py.

Referenced by PyPostYolo.PyPostYolo.freeze().

---

The documentation for this class was generated from the following file:

• share-jevoispro/pydnn/post/PyPostYolo.py