paarthn

4 ms·

1. 1.▲
   Natural Language and Computer Vision Intersection(paarthneekhara.github.io)

   3 points by paarthn 10y ago|0 comments
2. 2.▲
   Generating Videos with Scene Dynamics(github.com)

   5 points by paarthn 10y ago|0 comments
3. 3.▲
   by paarthn 10y ago

   I trained it on an AWS instance with grid k520 gpu.
4. 4.▲
   by paarthn 10y ago

   One possible improvement can be training the text embeddings along with the entire model (Instead of using the pretrained embeddings like skip-thought-vectors). It is on my to-do list, I'll try it out.
5. 5.▲
   by paarthn 10y ago

   There is still a long way to go, to be able to do that. The model currently generates 64 X 64 pictures and is trained on a very specific flowers image dataset. Nevertheless, it would be a great idea to experiment with such a dataset (of ske