Skip to content

RiDang/JMLN

Repository files navigation

JMLN: Joint Metric Learning Network for 2D Sketch-based 3D Shape Retrieval

Sketch-based 3D Shape Retrieval

Introduction

The code is used in the paper of JMLN, which attempts to further improve the cross-domain retrieval task. The framework is shown as followings:

image-20201019172033483

Environment

The code has been tested successfully on Ubuntu 18.04.3 LTS, Cuda 10.0, GTX 1080 Ti, Pytorch 1.4.0, Docker images ufoym/deepo:all-py36-cu100.

Note: Using docker is suggested.

Update

Data and Pre-trained Models

Training and Test Data

Please put related dataset into dataset/

Usage

Training and Test

# training 
sh train_all.sh

# Test after training
sh extract_all.sh

About

Joint Metric Learning Network for 2D Sketch-based 3D Shape Retrieval

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published