Dataset and assignment

.gitignore

@@ -30,3 +30,5 @@
 *.exe
 *.out
 *.app
+
+**/.DS_Store

Dataset/README.md

@@ -0,0 +1,43 @@
+# Sport video analysis for billiard matches
+
+
+## Overview
+The benchmark dataset consists of 10 different video clips of individual billiard shots extracted from 4 different match videos on YouTube. For each video clip, two frames were selected corresponding to the initial situation before the shot (i.e., “frame_first.png”) and the situation after the shot (i.e., “frame_last.png”).
+
+## Dataset structure
+
+The dataset is organized as the following. A folder is provided for each video clip, containing:
+
+- a clip video of a billiard shots (`gameX_clipY.mp4`);
+- a `frames` folder containing the frames extracted from the video clip for the initial status of the game (`frame_first.png`) and the final status of the game (`frame_last.png`);
+- a `bounding_boxes` folder containing the bounding box annotations of each image;
+- a `masks` folder containing the segmentation mask annotations of each image.
+
+
+## Annotation labels
+
+The benchmark dataset has been annotated according the following categories. Each category is assigned to a unique ID:
+
+0. Background
+1. white "cue ball"
+2. black "8-ball"
+3. ball with solid color
+4. ball with stripes
+5. playing field (table)
+
+
+The frames corresponding to the initial and final status of the game have been annotated with bounding boxes and segmentation masks:
+
+- Bounding boxes are defined for every ball in the image as a rectangle defined by 5 parameters [x, y, width, height, ball category ID], where (x,y) are the top-left corner coordinates and width and height are the bounding box main dimensions; the 5th parameter is the label ID representing the type of the ball; such parameters are listed in a row, one ball per row;
+
+- Segmentation masks are provided as grayscale mask where each pixel is assigned the corresponding category ID  (background, white cu ball, black 8-ball, solid color, striped and playing field).
+
+
+Note that segmentation masks can be easily visualized as color images highlighting the different categories by mapping each category ID in a segmentation mask to a RGB color, for example:
+
+0: (128,128,128)
+1: (255, 255, 255)
+2: (0,0,0)
+3: (0,0,255)
+4: (255,0,0,)
+5: (0,255,0)

Dataset/game1_clip1/bounding_boxes/frame_first_bbox.txt

@@ -0,0 +1,15 @@
+268 317 17 17 1
+469 154 15 15 2
+534 363 18 18 3
+321 324 18 18 3
+716 249 18 17 3
+417 221 18 17 3
+307 205 17 18 3
+469 186 17 17 3
+668 447 17 16 4
+444 396 18 18 4
+647 283 21 22 4
+535 251 18 19 4
+271 187 18 18 4
+834 186 18 18 4
+434 123 18 19 4

Dataset/game1_clip1/bounding_boxes/frame_last_bbox.txt

@@ -0,0 +1,15 @@
+321 201 17 17 1
+469 153 15 16 2
+535 364 16 16 3
+321 325 18 17 3
+717 249 17 17 3
+302 209 18 15 3
+469 186 17 17 3
+549 142 17 17 3
+668 446 16 17 4
+444 397 18 17 4
+650 284 17 17 4
+535 252 18 17 4
+272 187 16 17 4
+834 185 18 18 4
+434 123 18 19 4

Dataset/game1_clip2/bounding_boxes/frame_first_bbox.txt

@@ -0,0 +1,15 @@
+475 141 19 21 1
+696 223 22 23 2
+567 347 24 26 3
+399 252 22 23 3
+650 222 22 24 3
+600 200 21 23 3
+462 160 21 22 3
+617 156 19 22 3
+677 418 26 22 4
+262 318 24 26 4
+513 309 23 26 4
+560 252 23 26 4
+360 210 21 24 4
+733 209 21 22 4
+638 143 20 22 4

Dataset/game1_clip2/bounding_boxes/frame_last_bbox.txt

@@ -0,0 +1,15 @@
+620 161 23 22 1
+696 222 22 24 2
+567 346 24 27 3
+741 303 24 25 3
+399 251 22 24 3
+650 222 22 24 3
+462 161 21 21 3
+614 154 19 21 3
+677 418 26 23 4
+262 318 23 26 4
+511 309 27 28 4
+560 253 24 24 4
+360 210 21 24 4
+733 208 21 23 4
+638 142 20 22 4

Dataset/game1_clip3/bounding_boxes/frame_first_bbox.txt

@@ -0,0 +1,7 @@
+578 164 20 22 1
+540 282 23 25 2
+613 230 22 24 3
+309 189 21 22 3
+756 328 25 26 4
+429 328 24 27 4
+658 237 21 24 4

Dataset/game1_clip3/bounding_boxes/frame_last_bbox.txt

@@ -0,0 +1,6 @@
+306 284 23 24 1
+540 282 23 25 2
+309 188 21 23 3
+756 328 25 26 4
+429 328 24 26 4
+658 236 21 25 4

Dataset/game1_clip4/bounding_boxes/frame_first_bbox.txt

@@ -0,0 +1,14 @@
+347 334 24 26 1
+537 333 24 26 2
+409 382 25 27 3
+448 377 25 27 3
+614 245 22 24 3
+655 220 21 23 3
+566 146 20 22 3
+608 141 20 22 3
+687 292 23 24 4
+592 264 23 25 4
+358 220 22 24 4
+326 174 20 22 4
+547 112 20 21 4
+576 106 20 21 4

Dataset/game1_clip4/bounding_boxes/frame_last_bbox.txt

@@ -0,0 +1,14 @@
+425 360 25 26 1
+537 332 24 27 2
+794 420 25 24 3
+296 301 23 25 3
+614 245 22 24 3
+655 220 22 24 3
+566 146 20 22 3
+608 141 20 22 3
+687 292 23 25 4
+592 264 23 24 4
+358 220 22 23 4
+325 173 21 23 4
+547 112 20 21 4
+575 106 20 21 4

Dataset/game2_clip1/bounding_boxes/frame_first_bbox.txt

@@ -0,0 +1,12 @@
+740 259 25 24 1
+392 287 24 27 2
+820 473 31 12 3
+472 453 32 35 3
+437 210 21 24 3
+329 169 20 21 3
+374 147 18 20 3
+340 115 15 18 3
+378 92 18 16 3
+428 463 31 27 4
+903 461 34 25 4
+726 162 20 22 4

Dataset/game2_clip1/bounding_boxes/frame_last_bbox.txt

@@ -0,0 +1,12 @@
+359 150 19 19 1
+392 287 24 27 2
+821 473 30 13 3
+472 453 32 35 3
+437 210 21 24 3
+329 169 20 21 3
+365 115 17 17 3
+355 102 16 16 3
+378 91 17 16 3
+428 463 31 27 4
+904 461 32 26 4
+726 163 19 21 4

Dataset/game2_clip2/bounding_boxes/frame_first_bbox.txt

@@ -0,0 +1,13 @@
+527 119 18 18 1
+345 228 22 23 2
+758 292 24 26 3
+269 254 24 25 3
+342 195 20 22 3
+423 187 21 23 3
+578 100 17 18 3
+762 417 30 31 4
+337 349 28 31 4
+707 292 27 28 4
+733 238 23 25 4
+678 170 20 22 4
+328 144 18 20 4

Dataset/game2_clip2/bounding_boxes/frame_last_bbox.txt

@@ -0,0 +1,12 @@
+804 308 25 26 1
+423 188 20 21 2
+269 254 24 25 3
+345 227 22 24 3
+341 195 21 22 3
+578 100 17 18 3
+762 417 30 31 4
+337 348 28 32 4
+708 292 25 28 4
+733 238 23 25 4
+678 170 21 21 4
+328 143 18 20 4

Dataset/game3_clip1/bounding_boxes/frame_first_bbox.txt

@@ -0,0 +1,9 @@
+571 160 17 19 1
+604 248 19 21 2
+519 347 22 27 3
+377 315 24 25 3
+388 218 20 21 3
+406 322 23 24 4
+739 276 21 23 4
+694 263 22 22 4
+446 238 21 22 4

Dataset/game3_clip1/bounding_boxes/frame_last_bbox.txt

@@ -0,0 +1,8 @@
+455 306 22 24 1
+604 248 19 21 2
+519 347 23 25 3
+377 315 24 25 3
+406 321 23 25 4
+739 275 21 25 4
+694 263 22 25 4
+446 238 21 21 4

Dataset/game3_clip2/bounding_boxes/frame_first_bbox.txt

@@ -0,0 +1,11 @@
+695 244 22 22 1
+446 280 21 22 2
+451 328 23 24 3
+539 276 20 24 3
+539 221 19 22 3
+338 356 24 26 4
+279 328 23 25 4
+446 308 23 23 4
+503 179 19 19 4
+842 157 18 17 4
+540 122 17 17 4

Dataset/game3_clip2/bounding_boxes/frame_last_bbox.txt

@@ -0,0 +1,10 @@
+323 278 21 22 1
+446 280 21 22 2
+451 327 23 25 3
+539 276 20 24 3
+338 357 24 24 4
+279 328 23 24 4
+446 308 23 24 4
+504 179 18 19 4
+842 156 18 17 4
+539 122 18 17 4

Dataset/game4_clip1/bounding_boxes/frame_first_bbox.txt

@@ -0,0 +1,9 @@
+361 283 19 20 1
+457 270 17 20 2
+543 327 19 21 3
+713 359 21 21 4
+350 346 21 23 4
+469 336 19 20 4
+633 255 18 20 4
+621 91 16 16 4
+639 51 17 16 4

Dataset/game4_clip1/bounding_boxes/frame_last_bbox.txt

@@ -0,0 +1,8 @@
+302 292 18 20 1
+457 270 17 20 2
+713 359 21 22 4
+350 346 21 23 4
+469 334 19 22 4
+632 255 19 20 4
+621 90 15 17 4
+638 51 15 16 4

Dataset/game4_clip2/bounding_boxes/frame_first_bbox.txt

@@ -0,0 +1,13 @@
+466 237 18 19 1
+540 328 19 21 2
+698 254 17 18 3
+634 235 18 19 3
+479 208 17 19 3
+481 129 15 16 3
+560 114 16 17 3
+453 101 16 17 3
+560 52 13 15 3
+721 346 20 20 4
+528 304 19 20 4
+698 289 19 20 4
+618 170 16 18 4

Dataset/game4_clip2/bounding_boxes/frame_last_bbox.txt

@@ -0,0 +1,12 @@
+496 267 18 19 1
+540 328 19 21 2
+697 253 18 19 3
+634 235 18 19 3
+479 208 17 19 3
+481 128 16 17 3
+561 114 14 17 3
+453 101 16 17 3
+560 52 14 15 3
+527 304 20 21 4
+698 289 20 20 4
+617 170 18 18 4