Giters

akshitac8 / tfvaegan

[ECCV 2020] Official Pytorch implementation for "Latent Embedding Feedback and Discriminative Features for Zero-Shot Classification". SOTA results for ZSL and GZSL

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Recreating ResNet visual features

Pamulapati13 opened this issue · comments

commented

Hi @akshitac8 @naraysa ,

I was trying to recreate the ResNet visual features provided, from CUB dataset using images.
I used the pretrained ResNet from latest version of PyTorch.
It would be really helpful if you can help me to recreate the ResNet visual features from Images (for CUB dataset).

commented

@Pamulapati13 What problems are you facing?

commented

Hi @akshitac8

I was using pretrained ResNet 101 in pytorch to extract the visual features from individual images in CUB dataset.
What I found out was the visual features that I get from images in different from the visual features provided in the dataset(res101.mat file).
If you could share the details of ResNet model and preprocessing used to extract the features from images, that would be really helpful.

commented

@Pamulapati13 I used features provided by the original authors, which I have uploaded here https://drive.google.com/drive/u/3/folders/16Xk1eFSWjQTtuQivTogMmvL3P6F_084u
It would be really helpful if you could explain in detail what problems are you facing with the features?