norway vid

raw_data/videos/new_videos/commands.txt

@@ -1,7 +1,18 @@
+# ovaselu
+gdown 18GwrZ6pZBOB-21JIIxivcXfT6Soid2_M
 ffmpeg -i DJI_0372.MP4 -vf scale=960:540 -c:a copy ovaselu_DJI_0372_540p.mp4
+
+# politehnica
+gdown 15dzWjpPTybn9gQ-d-75woDxk2BIjJbXQ
 ffmpeg -i DJI_0741_a2.MP4 -vf scale=960:540 -c:a copy politehnica_DJI_0741_a2_540p.mp4
+
+# raciu
+gdown 1XYtq046nucfYTZSY06NQSqAMkHGJfNzh
 ffmpeg -i DJI_0418.MP4 -vf scale=960:540 -c:a copy raciu_DJI_0418_540p.mp4
-ffmpeg -i DJI_0372.MP4 -vf scale=960:540 -c:a copy norway_DJI_0372_540p.mp4
+
+# norway 0708
+gdown 1fTLXOfq4CouMIGGcMBG27nipajv5hbIz
+ffmpeg -i DJI_0708.MP4 -vf scale=960:540 -c:a copy norway_DJI_0708_540p.mp4
 
 # paris
 yt-dlp https://www.youtube.com/watch?v=rR_2c1YoEs8
@@ -15,5 +26,6 @@ ffmpeg -i 'San Francisco Via Drone Part 1 [Fsw1fWtyqrM].webm' -vf scale=960:540
 yt-dlp https://www.youtube.com/watch?v=1vjfHoNUNcE
 ffmpeg -i 'Rio De Janeiro Brazil via Drone [1vjfHoNUNcE].webm' -vf scale=960:540 -ss 00:00:01 -to 00:03:30 -c:v libx264 riodejaneiro_youtube_1_540p.mp4
 
+# sheerness dockyard
 yt-dlp https://www.youtube.com/watch?v=X-LFEO3XQhM
 ffmpeg -i 'Oil rig at Sheerness docks. [X-LFEO3XQhM].webm' -vf scale=960:540 -ss 00:00:25 -to 00:03:10 -c:v libx264 sheerness_youtube_1_540p.mp4

raw_data/videos/new_videos/{norway_DJI_0741_540p.mp4 → norway_DJI_0708_540p.mp4}

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:63467788015506f0f849f223a8aaae607caae97cfbf97bf143f1d71a5ed4d76e
-size 189178876
+oid sha256:44f9c14638bbb2d0037b0a4cefc5e62762bb4f017fcd789a45ddbd668cca81b4
+size 30453475

scripts/dronescapes_viewer/dronescapes_representations.py

@@ -13,7 +13,7 @@ def get_gt_tasks() -> dict[str, Representation]:
                 [255, 0, 0], [0, 0, 255], [0, 255, 255], [127, 127, 63]]
     classes_8 = ["land", "forest", "residential", "road", "little-objects", "water", "sky", "hill"]
     tasks = [
-        SemanticRepresentation("semantic_output", classes=classes_8, color_map=color_map),
+        SemanticRepresentation("semantic_output", classes=classes_8, color_map=color_map, disk_data_argmax=True),
         DepthRepresentation("depth_output", min_depth=0, max_depth=300),
         NormalsRepresentation("camera_normals_output"),
     ]
@@ -25,7 +25,7 @@ def get_other_tasks(include_semantics_original: bool, include_ci: bool) -> dict[
                                  BinaryMapper, mapillary_classes, coco_classes)
     tasks = [
         rgb := RGB("rgb"),
-        OpticalFlowRepresentation("opticalflow_rife"),
+        # OpticalFlowRepresentation("opticalflow_rife"),
         DepthRepresentation("depth_marigold", min_depth=0, max_depth=1),
         NormalsRepresentation("normals_svd(depth_marigold)")
     ]

scripts/m2f_metrics_analysis/m2f_main.ipynb

@@ -111,9 +111,9 @@
     "os.environ[\"VRE_DEVICE\"] = device = \"cuda\" #\"cpu\"\n",
     "os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"7\"\n",
     "\n",
-    "m2f_1 = Mask2Former(model_id, semantic_argmax_only=False, name=\"m2f\", dependencies=[])\n",
-    "m2f_2 = Mask2Former(\"47429163_0\", semantic_argmax_only=False, name=\"m2f\", dependencies=[])\n",
-    "m2f_3 = Mask2Former(\"49189528_1\", semantic_argmax_only=False, name=\"m2f\", dependencies=[])\n",
+    "m2f_1 = Mask2Former(model_id, disk_data_argmax=False, name=\"m2f\", dependencies=[])\n",
+    "m2f_2 = Mask2Former(\"47429163_0\", disk_data_argmax=False, name=\"m2f\", dependencies=[])\n",
+    "m2f_3 = Mask2Former(\"49189528_1\", disk_data_argmax=False, name=\"m2f\", dependencies=[])\n",
     "\n",
     "m2f_1.device = \"cuda\" if tr.cuda.is_available() else \"cpu\"\n",
     "m2f_2.device = \"cuda\" if tr.cuda.is_available() else \"cpu\"\n",

scripts/semantic_mapper/semantic_mapper.py

@@ -97,11 +97,11 @@ mapillary_color_map = [[165, 42, 42], [0, 192, 0], [196, 196, 196], [190, 153, 1
                        [0, 0, 70], [0, 0, 192], [32, 32, 32], [120, 10, 10]]
 
 m2f_coco = SemanticRepresentation("semantic_mask2former_coco_47429163_0", classes=coco_classes,
-                                  color_map=coco_color_map)
+                                  color_map=coco_color_map, disk_data_argmax=True)
 m2f_mapillary = SemanticRepresentation("semantic_mask2former_mapillary_49189528_0", classes=mapillary_classes,
-                                       color_map=mapillary_color_map)
+                                       color_map=mapillary_color_map, disk_data_argmax=True)
 m2f_r50_mapillary = SemanticRepresentation("semantic_mask2former_mapillary_49189528_1", classes=mapillary_classes,
-                                           color_map=mapillary_color_map)
+                                           color_map=mapillary_color_map, disk_data_argmax=True)
 marigold = DepthRepresentation("depth_marigold", min_depth=0, max_depth=1)
 normals_svd_marigold = NormalsRepresentation("normals_svd(depth_marigold)")
 
@@ -310,7 +310,7 @@ class SemanticMedian(TaskMapper, NpIORepresentation):
 
     @overrides
     def make_images(self, data: ReprOut) -> np.ndarray:
-        data_output = data.output.argmax(-1) if np.issubdtype(data.output.dtype, np.floating) else data.output
+        data_output = data.output.argmax(-1)# if np.issubdtype(data.output.dtype, np.floating) else data.output
         return colorize_semantic_segmentation(data_output, self.classes, self.color_map)
 
 class SafeLandingAreas(BinaryMapper, NpIORepresentation):

vre_dronescapes/cfg.yaml

@@ -30,7 +30,7 @@ representations:
     dependencies: []
     parameters:
       model_id: "47429163_0"
-      semantic_argmax_only: True
+      disk_data_argmax: True
     compute_parameters:
       batch_size: 1
 
@@ -39,7 +39,7 @@ representations:
     dependencies: []
     parameters:
       model_id: "49189528_0"
-      semantic_argmax_only: True
+      disk_data_argmax: True
     compute_parameters:
       batch_size: 1
 
@@ -48,7 +48,7 @@ representations:
     dependencies: []
     parameters:
       model_id: "49189528_1"
-      semantic_argmax_only: True
+      disk_data_argmax: True
     compute_parameters:
       batch_size: 1
 

vre_dronescapes/commands.txt

@@ -33,7 +33,7 @@ VRE_DEVICE=cuda CUDA_VISIBLE_DEVICES=7 vre ../raw_data/videos/olanesti_DJI_0416_
 # new videos
 
 tmux new -s norway2
-VRE_DEVICE=cuda CUDA_VISIBLE_DEVICES=1 vre ../raw_data/videos/new_videos/norway_DJI_0741_540p.mp4 -o norway_DJI_0741_540p/ --config_path cfg.yaml --output_dir_exists_mode skip_computed --exception_mode skip_representation --n_threads_data_storer 4 -I ../scripts/semantic_mapper/semantic_mapper.py:get_new_semantic_mapped_tasks
+VRE_DEVICE=cuda CUDA_VISIBLE_DEVICES=1 vre ../raw_data/videos/new_videos/norway_DJI_0708_540p.mp4 -o norway_DJI_0708_540p.mp4/ --config_path cfg.yaml --output_dir_exists_mode skip_computed --exception_mode skip_representation --n_threads_data_storer 4 -I ../scripts/semantic_mapper/semantic_mapper.py:get_new_semantic_mapped_tasks
 
 tmux new -s politehnica
 VRE_DEVICE=cuda CUDA_VISIBLE_DEVICES=2 vre ../raw_data/videos/new_videos/politehnica_DJI_0741_a2_540p.mp4 -o politehnica_DJI_0741_a2_540p/ --config_path cfg.yaml --output_dir_exists_mode skip_computed --exception_mode skip_representation --n_threads_data_storer 4 -I ../scripts/semantic_mapper/semantic_mapper.py:get_new_semantic_mapped_tasks
@@ -69,8 +69,8 @@ python scripts/symlinks_from_txt_list.py vre_dronescapes/ --copy_files --txt_fil
 
 # experts (make combined txt file of all new scenes + old data)
 # in vre_dronescapes_dir
-ls . | grep "norway_DJI_0741_540p\|politehnica_DJI_0741_a2_540p\|ovaselu_DJI_0372_540p\|raciu_DJI_0418_540p\|sanfrancisco_youtube_1_540p\|paris_youtube_1_540p\|riodejaneiro_youtube_1_540p\|sheerness_youtube_1_540p" | while read line; do rm -f ../scripts/txt_files/experts/"$line".txt; ls -v $line/rgb/npz | while read line2; do stem=${line2:0:-4}; echo $line/$stem >> ../scripts/txt_files/experts/"$line".txt; done; done
-ls . | grep "norway_DJI_0741_540p\|politehnica_DJI_0741_a2_540p\|ovaselu_DJI_0372_540p\|raciu_DJI_0418_540p\|sanfrancisco_youtube_1_540p\|paris_youtube_1_540p\|riodejaneiro_youtube_1_540p\|sheerness_youtube_1_540p" | while read line; do echo $line; cp -r $line/depth_marigold $line/depth_output; ../scripts/update_stastistics.py $line/depth_output/npz --statistics_file  ../data/train_set/.task_statistics.npz --source_task depth_marigold --n_workers 32; cp -r $line/'normals_svd(depth_marigold)' $line/camera_normals_output; done
+ls . | grep "norway_DJI_0708_540p.mp4\|politehnica_DJI_0741_a2_540p\|ovaselu_DJI_0372_540p\|raciu_DJI_0418_540p\|sanfrancisco_youtube_1_540p\|paris_youtube_1_540p\|riodejaneiro_youtube_1_540p\|sheerness_youtube_1_540p" | while read line; do rm -f ../scripts/txt_files/experts/"$line".txt; ls -v $line/rgb/npz | while read line2; do stem=${line2:0:-4}; echo $line/$stem >> ../scripts/txt_files/experts/"$line".txt; done; done
+ls . | grep "norway_DJI_0708_540p.mp4\|politehnica_DJI_0741_a2_540p\|ovaselu_DJI_0372_540p\|raciu_DJI_0418_540p\|sanfrancisco_youtube_1_540p\|paris_youtube_1_540p\|riodejaneiro_youtube_1_540p\|sheerness_youtube_1_540p" | while read line; do echo $line; cp -r $line/depth_marigold $line/depth_output; ../scripts/update_stastistics.py $line/depth_output/npz --statistics_file  ../data/train_set/.task_statistics.npz --source_task depth_marigold --n_workers 32; cp -r $line/'normals_svd(depth_marigold)' $line/camera_normals_output; done
 mkdir ../scripts/txt_files/experts # in root dir
 cd ../scripts/txt_files/experts # move to txt files dir
 cat ../annotated_and_segprop/train_files_11664.txt > train_set.txt