brettgohre / still_life_rendering_gqn

Generative Query Network trained on real still life photos. Implemented with Tensorflow.
MIT License
19 stars 2 forks source link

genrative query network #1

Open mjchen611 opened 6 years ago

mjchen611 commented 6 years ago

Hi Göhre,

Thank you for your great code of gqn in real image, I am a little curious about the following issues: How many epochs it use to train a model on real image? Can you show me an trained modle and a testing demo?

Thank you very much!

Best wishes,
Mingjia Chen

eduardohenriquearnold commented 5 years ago

I am also trying to reproduce the results.

I noticed that in the still dataset the viewpoints are usually the same ones. It would be very interesting to see how the model generalises to arbitrary viewpoints, like the videos provided in the supplementary material of the original paper.

Thanks.

WENDGOUNDI commented 4 years ago

Hi Göhre, Thank for sharing your code. I am trying to implement it but i am facing some errors (in the below image). In addition, to train with the still_fruit_dataset, you stated that < Replace gqn_tfr_provider.py with modified provider: __ python3 train_gqn_draw.py --data_dir /vol --dataset rooms_ring_camera --model_dir gqn --debug >, could you be more explicit? Thanks. Screenshot from 2019-12-19 19-10-52