# VTSCN

**Repository Path**: zzh26268/VTSCN

## Basic Information

- **Project Name**: VTSCN
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: MIT
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2025-06-30
- **Last Updated**: 2025-06-30

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Bridging Visual and Textual Semantics: Towards Consistency for Unbiased Scene Graph Generation

This is an official implementation for [paper](https://ieeexplore.ieee.org/abstract/document/10502321) "Bridging Visual and Textual Semantics: Towards Consistency for Unbiased Scene Graph Generation". 

Our code is on top of [SHA-GCL](https://github.com/dongxingning/SHA-GCL-for-SGG), we sincerely thank them for their well-designed codebase. You can refer to this link to build the basic environment and prepare the dataset.

[Pocket](https://github.com/fredzzhang/pocket) package is also required, please refer to this link to for necessary packages.

For the weighted predicate embedding, the vg-version can be downloaded in this [link](https://1drv.ms/u/c/60174365786eb250/EWO7Ow0xIcdBnQd4MSSVhVgBWpKej-8rBgJ6UZbV3Zdo-g?e=tFIVT6), the gqa-version can be downloaded in this [link](https://1drv.ms/u/c/60174365786eb250/EWO7Ow0xIcdBnQd4MSSVhVgBWpKej-8rBgJ6UZbV3Zdo-g?e=tFIVT6).

## Important
Our code implements VTSCN on MOTIFS, VCTree and VTransE.