Code Monkey home page Code Monkey logo

object-detection-evaluation-tool's Introduction

Object Detection evaluation tools

Object Detection evaluation tools

Introduction

Lots of object detection algorithms used mAp(mean Average precision) as evaluation metric, and yet their mAp calculation code embedded in algorithm implementation code, which is not easy to be separated. Here is offering an independent mAp calculation tool, you can do:

  • Calculating mAp for your object detection algorithm testing(11 points method supported)
  • Get recall and precision in any confidence threshold
  • Drawing PR curve
  • Drawing ROC curve(Done)
  • Get FPPI(False Positive Per Image) and FPPW(False Positive Per Window) in any threshold(Done)

Update

  • fix some bugs
  • sematic label name is supported now (e.g. "Person", "Car")

Requirements

  • python 2.X OR python 3.X
  • python-opencv (any version)
  • matplotlib 2.2.3

Input format

  • TXT format, see 'sample' folder
  • XML format, only Pascal VOC style supported

Usage

python test.py --dir ./sample/prediction ./sample/test_annos --ratio 0.5 --thre 0.7

Anything you want to know about usage can be found with typing:

python test.py --h

or

python test.py --help

object-detection-evaluation-tool's People

Contributors

wangzhe0623 avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar

object-detection-evaluation-tool's Issues

sample

大佬,例子是两个文件夹的大小不等 然后我删的只剩两个文件夹。算出来都是0 怎么回事呢
Processing label: Person
class Person Ap: 0
class Person precision: []
class Person recall: []
FPPW: inf FPPI inf
Processing label: Vehicle
class Vehicle Ap: 0
class Vehicle precision: []
class Vehicle recall: []
FPPW: inf FPPI inf
Processing label: Dryer
class Dryer Ap: 0
class Dryer precision: []
class Dryer recall: []
FPPW: inf FPPI inf
mAp: 0.0

test

大佬,为什么没有测评结果啊
2020-09-24 17-59-20 的屏幕截图

the format of TXT format,

@wangzhe0623 can you provide more information about the format of ground truth txt file and prediction txt file.
For ground truth txt file, I guess it is
For prediction txt file, is it ?

demo

大佬,跑你的demo,都有问题:
Num of prediction files: 729
Num of ground truth files: 721
groundtruths' size does not match predictions' size, please check ~
Traceback (most recent call last):
File "test.py", line 70, in
eval.run()
File "/home/112831/code/tools/Eval_object_detection-master/evaluation.py", line 272, in run
prediction_path, gt_path, predictions, groundtruths, file_format = self.load_all_files()
TypeError: cannot unpack non-iterable int object

no value

Hi @wangzhe0623 my class is 1 which means value will be 2 including backgroud i tried test.pybut get nothing as shown below.
ask

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.