Zhou Yu

How many iterations does the model need to reproduce the results? I use the same hyper-parameters as for the baseline faster rcnn w/o deform conv., the final loss is much...

the improvement of DCN over the baseline

1

Thanks for the DCN impl. in caffe. I am curious about the performance of the baseline model w/o deform_conv. Besides, have you tried more complex model such as ResNet-101?

evaluation results

1

Can this converted model reproduce exactly the same MAP score as that calculated in the original caffe framework?

question about the ILSVRC2015 labels

4

Where can I find the ILSVRC2015 labels？and how much accuracy gain can this operation have from using the ILSVRC 2012 labels?

how to generate the image based on the CLEVR-1.0 dataset w/o images

1

I downloaded the CLEVR-1.0 dataset without the images. Is is possible to use the scripts in this repo to generate the corresponding images?