-
Hello, thanks again for your work!
I have a question. In `unsupervised.py` I have `im1_s, im2_s` with shapes `(2, 320, 1152, ?)`. After first `downsample(im1_s, 4)` shapes became `(2, 80, 288, ?)`,…
-
Hi, thanks for your great work! It's really impressive that you do the segmentation translation from the virtual world to the real world scene. But when I run your code, I have a little problem with y…
-
Hi Simon!
Thanks for your work and code, it's very helpful to me!
But I have some questions now, please give me some advices.
I want to train your network on my own datasets, but the size of my dat…
-
```
class Classifier_Module(nn.Module):
def __init__(self, inplanes, dilation_series, padding_series, num_classes):
super(Classifier_Module, self).__init__()
self.conv2d_list =…
-
Hey Loic,
I am trying to train SPG on a 5-class labeled dataset (basically synthia: 360-degree-depthcams -> pointcloud)
I have the data in .ply format and as far as I can tell, loading of PC and…
-
Hi,
I only want to confirm the data/train on Kitti, so I would like to skip the synthia etc. download.
Is there a parameter in the config file or somewhere else to disable the other downloads?
…
-
Hi, I see that you perform mean normalization with:
mean = [104.920005, 110.1753, 114.785955]
stddev = 1 / 0.0039216
Are these values from ImageNet dataset? I'm curious to know if performing no…
-
Hi, thank you for the great work! I've successfully tried your trained model in ROS, it looks great!
Just a few questions to start training my own model:
1. Is that possible to provide more instru…
ghost updated
6 years ago
-
Hi, thanks for the wonderful code, it is really great work. I have several questions on the pre-training of Synthia.
1. I saw you only used 10 sequences from the dataset to do the pre-training, wh…
-
Hi,
in `losses.py`, forward-backward occlusion loss is implemented as:
```python
...
flow_diff_bw = flow_bw + flow_fw_warped
...
fb_occ_bw = tf.cast(length_sq(flow_diff_bw) > occ_thr…