morpheusthewhite

It is likely caused by some invalid image; trying to solve it means changing a lot of code, but I have not the time to do it. Since the choice...

> The training pictures are entered into the network in the order of the data set No, they are shuffled in a random manner. > tf.app.flags.DEFINE_integer('im_per_batch' 1,"Images to use per...

You just need to change the`_shuffle_roidb_inds()` in `lib/layer_utils/roi_data_layer.py`

I was referring in general to images which are not correctly processed in the algorithm, in your case images which have an empty sequence as `overlaps`

> Can I understand this? Invalid image means that there is no overlap between any anchors and the ground thruth. Yes, I'm almost sure it is so. > In your...

Checking system log (`journalctl`) it seems to be, as you thought, a problem of missing decoder For .webm videos ```log Warning: "No decoder available for type 'video/webm'." ``` While for...

I solved this by installing `qt-gstreamer` and `gst-libav` packages (I'm using Arch Linux). You'll probably need only the first one

With further tests I discovered that `qt-gstreamer` is needed for `.webm` videos, while both are needed for `.mp4` videos