Kyubyong Park

Hi @krasserm , Thanks for this project! I wonder if you have any plans to apply the faster-rcnn directly so we can input an image, not its prebuilt features.

That's very strange. The link is valid, but it returns an error. Too much traffic? I don't know. Anyway, sorry for your inconvenience. I'll find the solution.

Now the link is back. I think it was dead due to the traffic.

Have you downloaded the pretrained model file?

I'm consulting the TensorFlow.org homepage https://www.tensorflow.org/api_docs/python/

That's a very interesting question. I think it's positive. I have no experience on video sequence tasks, but seems each sequential frame of a video can be treated as an...

Thanks for your information. All sources above look great. My focus is to build free pronunication dictionaries for multiple languages, not just for English. The reason I prefer IPA to...

Thanks!

The encoder leaves artifacts, which are non-zeros, for paddings. It doesn't make sense, a query attends to those, so before applying softmax for getting score, I overwrote them with very...

Well, does that make any difference?