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
1.03k stars 228 forks source link

why didn't you mask pred_rel_pair_index in visualization code ? #142

Closed doda999 closed 2 years ago

doda999 commented 2 years ago

❓ Questions and Help

Hello.

About "2.visualize_SGDet.ipynb", I'd like to ask one question. In "get_info_by_index" function, you masked _pred_relscore and _pred_rellabel, using the threshold. Then I'm not sure why you didn't mask _pred_relpair as well, since I thought each relation pair in _pred_relpair corresponds to each relation label in _pred_rellabel.

Sorry if I misunderstood something, but I hope you would reply. Thank you.