SAM-Net Implementation

This is a PyTorch implementation of SAM-Net paper and can be used to reproduce the results in the paper.

This work focuses on detector-free image matching.

This repo contains training, evaluation and basic demo scripts used in our paper.

A large part of the code base is borrowed from the LoFTR Repository under its own separate license, terms and conditions.

Installation

conda env create -f environment.yaml
conda activate samnet

Get started

A demo to match one image pair is provided. To get a quick start,

Indoor: ./scripts/reproduce_test/indoor.sh
Outdoor: ./scripts/reproduce_test/outdoor.sh

Data Preparation

Please follow the training doc for data organization

Evaluation

1. ScanNet Evaluation

cd scripts/reproduce_test
bash indoor.sh

2. MegaDepth Evaluation

cd scripts/reproduce_test
bash outdoor.sh

Training

1. ScanNet Training

cd scripts/reproduce_train
bash indoor.sh

2. MegaDepth Training

cd scripts/reproduce_train
bash outdoor.sh

Useful data:

The extra files that you need to run SAM-Net can be found here.

• dump
• data
• pixloc
• assets