Iamanorange

Seems it's an old known issue: https://github.com/Riant/host-switch-plus/issues/31 https://github.com/Riant/host-switch-plus/issues/39

`make all`

Building graph should be put out of camera capture loop: ```python predictions = self.model(inputs, training_schedule) pred_flow = predictions['flow'] with tf.Session(config=config) as sess: saver.restore(blablabla) while ret: # Get frames and preprocessing...

You should feed 2 placeholders to self.model(), instead of 2 constants.

I haven't trained Flownet2. For inference: ```python # Build Graph input_a = tf.placeholder(dtype=tf.float32, shape=[1, None, None, 3]) input_b = tf.placeholder(dtype=tf.float32, shape=[1, None, None, 3]) inputs = { 'input_a': input_a, 'input_b':...

@t-naveenkumar @PiaCuk https://github.com/sampepose/flownet2-tf/issues/62

https://github.com/sampepose/flownet2-tf/issues/62

作者给的AWS挂掉了，我做了两个备份，你可以从这两个备份下载训练好的权重。 https://drive.google.com/file/d/1cft8EvnsBL5w4-REUeAaVWLeRx39hyHE/view?usp=sharing https://mega.nz/#!xk5DVYwB!hG2u_3ho8LWJl660Z5e0z-C0SlytB4SvlSkuoLEd568

Do not put `-DNDEBUG` inside `-std=c++11`.

Anywhere. Just do NOT put one flag insider another flag.