Evaluating context-sensitive word meaning understanding in pair of sentences for BERT and GloVe+BiLSTM using WiC dataset | A3 for COL772 course (Fall 21)
-
Updated
Dec 5, 2021 - Jupyter Notebook
Evaluating context-sensitive word meaning understanding in pair of sentences for BERT and GloVe+BiLSTM using WiC dataset | A3 for COL772 course (Fall 21)
SuperGlue training pipeline written using Pytorch Lightning
Coarse-to-Fine High Resolution Image Alignment/Registration
Model Evaluation using SuperGLUE Diagnostic Dataset
image stitching using superglue
A PyTorch implementation of SuperGlue
This repository consists of two major components for a project called E-waste recycling system. One component is used to generate a labelled dataset to train U-Net segmentation network. The second component is used to perform non-rigid registration using Demon's algorithm.
This project uses the model from the paper SuperGlue: Learning Feature Matching with Graph Neural Networks and builds on it to implement an analysis of the images returned by the Hand-Up Display. In this project, significant results have been achieved on challenging "complex" geometric images.
The implementation about feature matching using various method !!
GPS기반 온라인 보물찾기 게임 서비스입니다
This project involves analyzing and classifying the BoolQ dataset from the SuperGLUE benchmark. We implemented various classifiers and techniques, including rules-based logic, BERT, RNN, and GPT-3/4 data augmentation, achieving performance improvements.
Source code for COMP90086 Project 2021
simple library to make life easy when deploying superpoint, superglue models
🦆 A simple sinkhorn algorithm to solve optimal transport problem writen in Matlab
CVPR 2022 "Image Matching: Local Features and Beyond" workshop challenge: Kaggle Silver Medal solution (34th out of 642 teams).
Benchmarking and evaluation framework for place recognition methods, featuring SuperPoint+SuperGlue, LoGG3D-Net, Scan Context, DBoW2, MixVPR, STD
Another way to not go to space today! What can be more Kerbal than more Explosions!* *Now with SuperKlue Implosions!* KaboOom! (BOOM) is an add-on for Kerbal Space Program.
scripted and traced deep learning models and their applications
A PyTorch Lightning extension that accelerates and enhances foundation model experimentation with flexible fine-tuning schedules.
Add a description, image, and links to the superglue topic page so that developers can more easily learn about it.
To associate your repository with the superglue topic, visit your repo's landing page and select "manage topics."