PI-DECODER

Introduction

PI-DECODER, a decoder structure designed for Precise Iris Segmentation and Location. The decoder structure is shown below:

Please check technical paper.pdf in the "reference" subfolder for more details.

How to use?

For african dataset, you can enter the following script on your terminal:

python main.py --mode test --model_path ./models/african_best.pth --test_mode 1 --train_dataset african

Then you have iris mask, pupil mask and outer iris mask that are predicted by the input images. At the same time, the relevant index data will be displayed on your terminal.

(ijcb) PS F:\workspace\code\pytorch\PI-DECODER> python main.py --mode test --model_path ./models/african_best.pth --
test_mode 1 --train_dataset african
Namespace(batch_size=1, beta1=0.9, beta2=0.999, img_size=(640, 640), lr=0.0002, mode='test', model_path='./models/af
rican_best.pth', num_epochs=100, num_workers=2, result_path='./result/', test_mode=1, test_path='./dataset/test/', t
rain_dataset='african', train_path='./dataset/train/', valid_path='./dataset/valid/')
image count in train path :5
image count in valid path :5
image count in test path :40
Using Model: PI-DECODER
0.0688 seconds per image

----------------------------------------------------------------------------------------------------------------
|evaluation     |e1(%)          |e2(%)          |miou(%)        |f1(%)          |miou_back      |f1_back        |
----------------------------------------------------------------------------------------------------------------
|iris seg       |0.384026       |0.192013       |91.175200      |95.350625      |95.386805      |97.574698      |
|iris mask      |0.569627       |0.284813       |93.159855      |96.430411      |96.270919      |98.060105      |
|pupil mask     |0.078793       |0.039396       |93.138878      |96.409347      |96.529547      |98.184718      |
----------------------------------------------------------------------------------------------------------------
|average        |0.344149       |0.172074       |92.491311      |96.063461      |96.062424      |97.939840      |
----------------------------------------------------------------------------------------------------------------

Besides, if you don't have groud-truth files or just want to save the results, use test mode 2.

python main.py --mode test --model_path ./models/african_best.pth --test_mode 2 --train_dataset african

Requirements

The whole experiment was run on the NVIDIA RTX 3060. The following are recommended environment configurations.

matplotlib        3.3.4
numpy             1.19.5
opencv-python     4.5.1.48
pandas            1.1.5
Pillow            8.1.2
pip               21.0.1
pyparsing         2.4.7
python-dateutil   2.8.1
pytz              2021.1
scipy             1.5.4
setuptools        52.0.0.post20210125
six               1.15.0
thop              0.0.31.post2005241907
torch             1.7.0+cu110
torchstat         0.0.7
torchsummary      1.5.1
torchvision       0.8.1+cu110

precise iris segmentation

Related tags

Overview

PI-DECODER

Introduction

How to use?

Requirements

Owner

Python interface for converting Penn Treebank trees to Stanford Dependencies and Universal Depenencies

Labelling platform for text using distant supervision

Pangu-Alpha for Transformers

Graph4nlp is the library for the easy use of Graph Neural Networks for NLP

Composed Image Retrieval using Pretrained LANguage Transformers (CIRPLANT)

Finding Label and Model Errors in Perception Data With Learned Observation Assertions

Correctly generate plurals, ordinals, indefinite articles; convert numbers to words

Search-Engine - 📖 AI based search engine

PyJPBoatRace: Python-based Japanese boatrace tools 🚤

An open-source NLP library: fast text cleaning and preprocessing.

Graph Coloring - Weighted Vertex Coloring Problem

Code and data accompanying Natural Language Processing with PyTorch

String Gen + Word Checker

open-information-extraction-system, build open-knowledge-graph(SPO, subject-predicate-object) by pyltp(version==3.4.0)

Trankit is a Light-Weight Transformer-based Python Toolkit for Multilingual Natural Language Processing

This repository will contain the code for the CVPR 2021 paper "GIRAFFE: Representing Scenes as Compositional Generative Neural Feature Fields"

BROS: A Pre-trained Language Model Focusing on Text and Layout for Better Key Information Extraction from Documents

Materials (slides, code, assignments) for the NYU class I teach on NLP and ML Systems (Master of Engineering).

Guide to using pre-trained large language models of source code

An implementation of WaveNet with fast generation