updated readme.

README.md

@@ -4,6 +4,46 @@ This dataset is an extension of the original [dronescapes dataset](https://huggi
 
 ![Logo](logo.png)
 
-WIP
+# 1. Downloading the data
 
+## Option 1. Download the pre-processed dataset from HuggingFace repository
 
+```
+git lfs install # Make sure you have git-lfs installed (https://git-lfs.com)
+git clone https://huggingface.co/datasets/Meehai/dronescapes
+```
+
+## Option 2. Generate all the modalities from raw videos
+
+Follow the instructions under [this file](./vre_dronescapes/commands.txt).
+
+Note: you can generate all the data except `semantic_segprop8` (human annotated), `depth_sfm_manual202204` and
+`normals_sfm_manual202204` (SfM tool was used).
+
+## 2. Using the data
+
+As per the split from the paper:
+
+<details>
+<summary> Split </summary>
+<img src="split.png">
+</details>
+
+The data is in `data/*` (if you used git clone) (it should match even if you download from huggingface).
+
+## 2.1 Using the provided viewer
+
+![Collage](collage.png)
+
+The simplest way to explore the data is to use the [provided notebook](scripts/dronescapes_viewer.ipynb). Upon running
+it, you should get a collage with all the default tasks, like the picture at the top.
+
+For a CLI-only method, you can use the provided reader as well:
+
+```
+python scripts/dronescapes_viewer.py data/test_set_annotated_only/ # or any of the 8 directories in data/
+```
+
+## 3. Evaluation
+
+See the original [dronescapes evaluation description & benchmark](https://huggingface.co/datasets/Meehai/dronescapes#3-evaluation-for-semantic-segmentation) for this.

vre_dronescapes/commands.txt

@@ -1,5 +1,6 @@
-# rgb semantic_mask2former_mapillary_49189528_0 semantic_mask2former_swin_mapillary_converted depth_marigold semantic_mask2former_swin_mapillary_converted semantic_mask2former_swin_coco_converted opticalflow_rife normals_svd(depth_marigold)
+# Run VRE on the raw videos
 
+# rgb semantic_mask2former_mapillary_49189528_0 semantic_mask2former_swin_mapillary_converted depth_marigold semantic_mask2former_swin_mapillary_converted semantic_mask2former_swin_coco_converted opticalflow_rife normals_svd(depth_marigold)
 tmux new -s vre_atanasie
 VRE_DEVICE=cuda CUDA_VISIBLE_DEVICES=0 vre ../raw_data/videos/atanasie_DJI_0652_full/atanasie_DJI_0652_full_540p.mp4 -o atanasie_DJI_0652_full/ --config_path cfg.yaml --output_dir_exists_mode skip_computed --representations rgb semantic_mask2former_mapillary_49189528_0 semantic_mask2former_swin_mapillary_converted --n_threads_data_storer 2 -I semantic_mapper.py:get_new_semantic_mapped_tasks;
 
@@ -25,3 +26,14 @@ VRE_DEVICE=cuda CUDA_VISIBLE_DEVICES=6 vre ../raw_data/videos/slanic_DJI_0956_09
 tmux new -s vre_olanesti_jupiter
 VRE_DEVICE=cuda CUDA_VISIBLE_DEVICES=7 vre ../raw_data/videos/jupiter_DJI_0703_0704_0705_combined_sliced_10650_21715/jupiter_DJI_0703_0704_0705_combined_sliced_10650_21715_540p.mp4 -o jupiter_DJI_0703_0704_0705_combined_sliced_10650_21715/ --config_path cfg.yaml --output_dir_exists_mode skip_computed --representations rgb semantic_mask2former_mapillary_49189528_0 semantic_mask2former_swin_mapillary_converted --n_threads_data_storer 2 -I semantic_mapper.py:get_new_semantic_mapped_tasks
 VRE_DEVICE=cuda CUDA_VISIBLE_DEVICES=7 vre ../raw_data/videos/olanesti_DJI_0416_full/olanesti_DJI_0416_full_540p.mp4 -o olanesti_DJI_0416_full/ --config_path cfg.yaml --output_dir_exists_mode skip_computed --representations rgb semantic_mask2former_mapillary_49189528_0 semantic_mask2former_swin_mapillary_converted --n_threads_data_storer 2 -I semantic_mapper.py:get_new_semantic_mapped_tasks
+
+# Split dataset in the relevant 8 splits
+
+python scripts/symlinks_from_txt_list.py vre_dronescapes/ --copy_files --txt_file txt_files/annotated_and_segprop/train_files_11664.txt -o data/train_set
+python scripts/symlinks_from_txt_list.py vre_dronescapes/ --copy_files --txt_file txt_files/annotated_and_segprop/val_files_605.txt -o data/validation_set
+python scripts/symlinks_from_txt_list.py vre_dronescapes/ --copy_files --txt_file txt_files/annotated_and_segprop/semisup_files_11299.txt -o data/semisupervised_set
+python scripts/symlinks_from_txt_list.py vre_dronescapes/ --copy_files --txt_file txt_files/annotated_and_segprop/test_files_5603.txt -o data/test_set
+python scripts/symlinks_from_txt_list.py vre_dronescapes/ --copy_files --txt_file txt_files/manually_annotated_files/train_files_218.txt -o data/train_set_annotated_only
+python scripts/symlinks_from_txt_list.py vre_dronescapes/ --copy_files --txt_file txt_files/manually_annotated_files/val_files_15.txt -o data/validation_set_annotated_only
+python scripts/symlinks_from_txt_list.py vre_dronescapes/ --copy_files --txt_file txt_files/manually_annotated_files/semisup_files_207.txt -o data/semisupervised_set_annotated_nly
+python scripts/symlinks_from_txt_list.py vre_dronescapes/ --copy_files --txt_file txt_files/manually_annotated_files/test_files_116.txt -o data/test_set_annotated_only