Oscar icon indicating copy to clipboard operation
Oscar copied to clipboard

coco datasets

Open santon834 opened this issue 4 years ago • 4 comments

Hi,

What is the relationship between the following datasets and how can i get from one to the other?

https://biglmdiag.blob.core.windows.net/vinvl/image_features/coco_X152C4_frcnnbig2_exp168model_0060000model.roi_heads.nm_filter_2_model.roi_heads.score_thresh_0.2/model_0060000/

https://biglmdiag.blob.core.windows.net/vinvl/datasets/coco_caption

Thanks, Anton

santon834 avatar Apr 28 '21 00:04 santon834

The above only contains image feature. The below is for coco-caption experiments, It can be used for reproducing the results reported in the paper. containing experiment configs, training/val/test image features, caption labels and so on.

fortunechen avatar May 13 '21 08:05 fortunechen

hi, Do you know how the author generated the data provided in the link below(https://biglmdiag.blob.core.windows.net/vinvl/datasets/coco_caption)?

ckmstydy avatar Jul 28 '21 03:07 ckmstydy

hi, Do you know how the author generated the data provided in the link below(https://biglmdiag.blob.core.windows.net/vinvl/datasets/coco_caption)?

I don't know, maybe you should open a new issue

fortunechen avatar Jul 28 '21 04:07 fortunechen

hi, Do you know how the author generated the data provided in the link below(https://biglmdiag.blob.core.windows.net/vinvl/datasets/coco_caption)?

Hi ckmstydy, You can generate input features from any folder of .jpg images with these steps.

When you're done with generating predictions.tsv, you use this comment to generate the additional 6 features needed, before you run the VinVL infererence.

There's some steps that has to be done before running the model on your images.. Good luck!

jontooy avatar Oct 19 '21 12:10 jontooy