Vkitti - Githubissues

Add pretraining of the seg and depth heads on vkitti 2

Data

data is downloaded and extracted in /miniscratch/_groups/ccai/data/vkitti2 (with an about.txt of course!)
copied all exising jsons to /miniscratch/_groups/ccai/data/jsons
put train_kitti.json and val_kitti.json there
dataset is described there -> https://europe.naverlabs.com/research/computer-vision/proxy-virtual-worlds-vkitti-2/
selected the entire Scene02 as validation data
Removed the rain sunset fog and morning variations

added data.py/process_kitti_seg() function to parse and merge labels from the rgb png images of the segmentation maps
depth is encoded as 1 px = 1cm so I just normalize the inverse depth in tutils.py/get_normalized_depth_t()
input image is loaded as a regular jpg image

I changed jsut a little the trainer.py/setup() interface with self.all_loaders, self.kitty_display_images and self.base_display_images
Added a simple utility in the trainer switch_data(self, to="kitti") or switch_data(self, to="base") that changes self.loaders and self.display_images to point to either the kitti data or the base data we are used to
NOTHING changes in the rest of the code :)
- Except that we need to catch here and there potential mismatches between tasks and data keys (when pretraining with kitti, tasks are msd but in the pre-training phase there is no m data)