search

KaihuaTang / Scene-Graph-Benchmark.pytorch

A new codebase for popular Scene Graph Generation methods (2020). Visualization & Scene Graph Extraction on custom images/datasets are provided. It's also a PyTorch implementation of paper “Unbiased Scene Graph Generation from Biased Training CVPR 2020”
MIT License
1.06k stars 228 forks source link

Transformer model details #96

Open coldmanck opened 3 years ago

coldmanck commented 3 years ago

❓ Questions and Help

Hi @KaihuaTang, thanks for your awesome work! I am wondering where I can find the details or an illustration of the Transformer model, as I couldn't find it anywhere in this codebase and the (Unbiased SGG) paper. Do I miss any description about it?

KaihuaTang commented 3 years ago

The code is here https://github.com/KaihuaTang/Scene-Graph-Benchmark.pytorch/blob/master/maskrcnn_benchmark/modeling/roi_heads/relation_head/model_transformer.py

In the paper, we don't use the transformer, since we need to improve "existing" methods.

coldmanck commented 3 years ago

Thanks for your prompt reply! I understand the model is inside this code base, I was wondering is there any explanation about the architecture.