JeVoisBase  1.20
JeVois Smart Embedded Machine Vision Toolkit Base Modules
Share this page:
PyPostDAMOyolo.PyPostDAMOyolo Class Reference

Python DNN post-processor for DAMO YOLO. More...

Public Member Functions

def __init__ (self)
 Constructor. More...
 
def init (self)
 JeVois parameters initialization. More...
 
def freeze (self, doit)
 Freeze some parameters that should not be changed at runtime. More...
 
def loadClasses (self, filename)
 Parameter callback: Load class names when 'classes' parameter value is changed by model zoo. More...
 
def multiclass_nms (self, bboxes, scores, score_th, nms_th, max_num)
 Multiclass non-maximum suppression as implemented by PINTO0309. More...
 
def postprocess (self, scores, bboxes, score_th, nms_th)
 Post-processing as implemented by PINTO0309. More...
 
def process (self, outs, preproc)
 Get network outputs outs is a list of numpy arrays for the network's outputs. 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)
 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
 
 classes
 
 nms
 
 maxnbox
 
 cthresh
 

Detailed Description

Python DNN post-processor for DAMO YOLO.

Adapted from https://github.com/PINTO0309/PINTO_model_zoo/blob/main/334_DAMO-YOLO/demo/demo_DAMO-YOLO_onnx.py and https://github.com/tinyvision/DAMO-YOLO/blob/master/tools/demo.py

Author
Laurent Itti
Email:
itti@usc.edu
Address:
University of Southern California, HNB-07A, 3641 Watt Way, Los Angeles, CA 90089-2520, USA
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 25 of file PyPostDAMOyolo.py.

Constructor & Destructor Documentation

◆ __init__()

def PyPostDAMOyolo.PyPostDAMOyolo.__init__ (   self)

Constructor.

Definition at line 28 of file PyPostDAMOyolo.py.

Member Function Documentation

◆ freeze()

def PyPostDAMOyolo.PyPostDAMOyolo.freeze (   self,
  doit 
)

Freeze some parameters that should not be changed at runtime.

Definition at line 61 of file PyPostDAMOyolo.py.

References PyPostClassify.PyPostClassify.classes, PyPostDAMOyolo.PyPostDAMOyolo.classes, Darknet.classes, and Yolo.classes.

◆ getLabel()

def PyPostDAMOyolo.PyPostDAMOyolo.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 149 of file PyPostDAMOyolo.py.

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

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

◆ init()

def PyPostDAMOyolo.PyPostDAMOyolo.init (   self)

JeVois parameters initialization.

These can be set by users in the GUI or JeVois model zoo file

Definition at line 39 of file PyPostDAMOyolo.py.

◆ loadClasses()

def PyPostDAMOyolo.PyPostDAMOyolo.loadClasses (   self,
  filename 
)

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

Definition at line 66 of file PyPostDAMOyolo.py.

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

◆ multiclass_nms()

def PyPostDAMOyolo.PyPostDAMOyolo.multiclass_nms (   self,
  bboxes,
  scores,
  score_th,
  nms_th,
  max_num 
)

Multiclass non-maximum suppression as implemented by PINTO0309.

Definition at line 74 of file PyPostDAMOyolo.py.

Referenced by PyPostDAMOyolo.PyPostDAMOyolo.postprocess().

◆ postprocess()

def PyPostDAMOyolo.PyPostDAMOyolo.postprocess (   self,
  scores,
  bboxes,
  score_th,
  nms_th 
)

◆ process()

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

Get network outputs 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 113 of file PyPostDAMOyolo.py.

References PyPostDAMOyolo.PyPostDAMOyolo.boxes, PyPostDAMOyolo.PyPostDAMOyolo.classIds, PyPostDAMOyolo.PyPostDAMOyolo.confidences, PyPostClassify.PyPostClassify.cthresh, PyPostDAMOyolo.PyPostDAMOyolo.cthresh, demo.float, PyPostDAMOyolo.PyPostDAMOyolo.nms, and PyPostDAMOyolo.PyPostDAMOyolo.postprocess().

◆ report()

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

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 168 of file PyPostDAMOyolo.py.

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

Member Data Documentation

◆ boxes

◆ classes

◆ classIds

◆ classmap

◆ confidences

◆ cthresh

PyPostDAMOyolo.PyPostDAMOyolo.cthresh

Definition at line 55 of file PyPostDAMOyolo.py.

Referenced by PyPostDAMOyolo.PyPostDAMOyolo.process().

◆ maxnbox

PyPostDAMOyolo.PyPostDAMOyolo.maxnbox

Definition at line 51 of file PyPostDAMOyolo.py.

Referenced by PyPostDAMOyolo.PyPostDAMOyolo.postprocess().

◆ nms

PyPostDAMOyolo.PyPostDAMOyolo.nms

Definition at line 47 of file PyPostDAMOyolo.py.

Referenced by PyPostDAMOyolo.PyPostDAMOyolo.process().


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