luizgh/sigver_wiwd
Learned representation for Offline Handwritten Signature Verification. Models and code to extract features from signature images.
Pre-trained CNN models extract fixed-size feature vectors (2048-dim) from variable-sized signature images using SigNet and SigNet-SPP architectures, with the latter employing spatial pyramid pooling to handle signatures of different resolutions. Built on Theano/Lasagne with TensorFlow compatibility, the framework includes image preprocessing pipelines, batch processing utilities, and pre-extracted feature datasets for GPDS, MCYT, CEDAR, and Brazilian PUC-PR benchmarks to support offline signature verification research.
156 stars. No commits in the last 6 months.
Stars
156
Forks
52
Language
Jupyter Notebook
License
BSD-2-Clause
Category
Last pushed
Feb 22, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/luizgh/sigver_wiwd"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
Aftaab99/OfflineSignatureVerification
Writer independent offline signature verification using convolutional siamese networks
anupdhoble/6thSemProject_SignatureVerification
This repository contains the source code and documentation for a Signature Verification System...
muame-amr/Signature-Verification
Signature Verification Convolutional Neural Network (CNN) model using Deep Leaning with...
sainipankaj15/Signature-Forgery-Detection
Here I am adding all files and code which I used during project. I also explained project in below video
TranHuuDat2004/handwriting-signature-recognition
Developed a machine learning model for authenticating handwritten signatures. This project...