Extract own image features

KaihuaTang / VQA2.0-Recent-Approachs-2018.pytorch

A pytroch reimplementation of "Bilinear Attention Network", "Intra- and Inter-modality Attention", "Learning Conditioned Graph Structures", "Learning to count object", "Bottom-up top-down" for Visual Question Answering 2.0

GNU General Public License v3.0

295 stars 57 forks source link

Extract own image features #10

Open dimvarsamis opened 3 years ago

dimvarsamis commented 3 years ago

Hello, this is a really nice work, congratulations!! I wanted to ask if is it possible to extract my own image features during the train process of the model and not load them.

thanks