search

douglasrizzo / JINT2020-ball-detection

Code and data from our 2020 JINT paper
MIT License
9 stars 5 forks source link

./darknet detector map .... --> All #1

Open giorgiomondauto opened 3 years ago

giorgiomondauto commented 3 years ago

Hey Great Repo. Thanks for it. One question: Do you also have the results from your last command? I followed all the steps as listed in the repo and I am running the ./darknet detector map ....

but I am getting all FNegatives and detections_count = 0. Is there something I am missing? I am using the finalweights in here: https://github.com/douglasrizzo/JINT2020-ball-detection/tree/master/networks/yolo/weights_trained

douglasrizzo commented 3 years ago

Hello there. I just re-ran one experiment with a YOLOv3 model. The results of the command are below. I had some trouble downloading the weights file from this repo, because I used Git LFS to upload the files (they are very large) and apparently I have "exceeded my storage quota" or something, so I suggest you take a look if your weights files are all intact.

If not, I can provide you with a download link so you can test them with the weights file I used in this example (should be the same).

~/Code/darknet/darknet detector map JINT2020-ball-detection/data/yolo/obj.data JINT2020-ball-detection/networks/yolo/configs/yolov3.cfg JINT2020-ball-detection/networks/yolo/weights_trained/yolov3_final.weights -points 0
 GPU isn't used
 Used AVX
 Used FMA & AVX2
 OpenCV version: 4.5.0
mini_batch = 1, batch = 1, time_steps = 1, train = 0
   layer   filters  size/strd(dil)      input                output
   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF
   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF
   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF
   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF
   4 Shortcut Layer: 1,  wt = 0, wn = 0, outputs: 208 x 208 x  64 0.003 BF
   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF
   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF
   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF
   8 Shortcut Layer: 5,  wt = 0, wn = 0, outputs: 104 x 104 x 128 0.001 BF
   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF
  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF
  11 Shortcut Layer: 8,  wt = 0, wn = 0, outputs: 104 x 104 x 128 0.001 BF
  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF
  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF
  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF
  15 Shortcut Layer: 12,  wt = 0, wn = 0, outputs:  52 x  52 x 256 0.001 BF
  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF
  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF
  18 Shortcut Layer: 15,  wt = 0, wn = 0, outputs:  52 x  52 x 256 0.001 BF
  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF
  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF
  21 Shortcut Layer: 18,  wt = 0, wn = 0, outputs:  52 x  52 x 256 0.001 BF
  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF
  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF
  24 Shortcut Layer: 21,  wt = 0, wn = 0, outputs:  52 x  52 x 256 0.001 BF
  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF
  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF
  27 Shortcut Layer: 24,  wt = 0, wn = 0, outputs:  52 x  52 x 256 0.001 BF
  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF
  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF
  30 Shortcut Layer: 27,  wt = 0, wn = 0, outputs:  52 x  52 x 256 0.001 BF
  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF
  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF
  33 Shortcut Layer: 30,  wt = 0, wn = 0, outputs:  52 x  52 x 256 0.001 BF
  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF
  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF
  36 Shortcut Layer: 33,  wt = 0, wn = 0, outputs:  52 x  52 x 256 0.001 BF
  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF
  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF
  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF
  40 Shortcut Layer: 37,  wt = 0, wn = 0, outputs:  26 x  26 x 512 0.000 BF
  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF
  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF
  43 Shortcut Layer: 40,  wt = 0, wn = 0, outputs:  26 x  26 x 512 0.000 BF
  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF
  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF
  46 Shortcut Layer: 43,  wt = 0, wn = 0, outputs:  26 x  26 x 512 0.000 BF
  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF
  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF
  49 Shortcut Layer: 46,  wt = 0, wn = 0, outputs:  26 x  26 x 512 0.000 BF
  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF
  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF
  52 Shortcut Layer: 49,  wt = 0, wn = 0, outputs:  26 x  26 x 512 0.000 BF
  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF
  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF
  55 Shortcut Layer: 52,  wt = 0, wn = 0, outputs:  26 x  26 x 512 0.000 BF
  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF
  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF
  58 Shortcut Layer: 55,  wt = 0, wn = 0, outputs:  26 x  26 x 512 0.000 BF
  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF
  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF
  61 Shortcut Layer: 58,  wt = 0, wn = 0, outputs:  26 x  26 x 512 0.000 BF
  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF
  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF
  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF
  65 Shortcut Layer: 62,  wt = 0, wn = 0, outputs:  13 x  13 x1024 0.000 BF
  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF
  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF
  68 Shortcut Layer: 65,  wt = 0, wn = 0, outputs:  13 x  13 x1024 0.000 BF
  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF
  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF
  71 Shortcut Layer: 68,  wt = 0, wn = 0, outputs:  13 x  13 x1024 0.000 BF
  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF
  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF
  74 Shortcut Layer: 71,  wt = 0, wn = 0, outputs:  13 x  13 x1024 0.000 BF
  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF
  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF
  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF
  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF
  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF
  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF
  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF
  82 yolo
[yolo] params: iou loss: mse (2), iou_norm: 0.75, obj_norm: 1.00, cls_norm: 1.00, delta_norm: 1.00, scale_x_y: 1.00
  83 route  79                                 ->   13 x  13 x 512
  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF
  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256
  86 route  85 61                              ->   26 x  26 x 768
  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF
  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF
  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF
  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF
  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF
  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF
  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF
  94 yolo
[yolo] params: iou loss: mse (2), iou_norm: 0.75, obj_norm: 1.00, cls_norm: 1.00, delta_norm: 1.00, scale_x_y: 1.00
  95 route  91                                 ->   26 x  26 x 256
  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF
  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128
  98 route  97 36                              ->   52 x  52 x 384
  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF
 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF
 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF
 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF
 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF
 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF
 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF
 106 yolo
[yolo] params: iou loss: mse (2), iou_norm: 0.75, obj_norm: 1.00, cls_norm: 1.00, delta_norm: 1.00, scale_x_y: 1.00
Total BFLOPS 65.304
avg_outputs = 516723
Loading weights from JINT2020-ball-detection/networks/yolo/weights_trained/yolov3_final.weights...
 seen 64, trained: 384 K-images (6 Kilo-batches_64)
Done! Loaded 107 layers from weights-file

 calculation mAP (mean average precision)...
 Detection layer: 82 - type = 28
 Detection layer: 94 - type = 28
 Detection layer: 106 - type = 28
252
 detections_count = 414, unique_truth_count = 250
class_id = 0, name = ball, ap = 13.95%       (TP = 49, FP = 207)

 for conf_thresh = 0.25, precision = 0.19, recall = 0.20, F1-score = 0.19
 for conf_thresh = 0.25, TP = 49, FP = 207, FN = 201, average IoU = 14.22 %

 IoU threshold = 50 %, used Area-Under-Curve for each unique Recall
 mean average precision (mAP@0.50) = 0.139467, or 13.95 %
Total Detection Time: 191 Seconds

Set -points flag:
 `-points 101` for MS COCO
 `-points 11` for PascalVOC 2007 (uncomment `difficult` in voc.data)
 `-points 0` (AUC) for ImageNet, PascalVOC 2010-2012, your custom dataset
giorgiomondauto commented 3 years ago

Hi Douglas, thanks for your quick reply. I think you may be right, the problem should be related to the weights. If you could send me the link to download the weights, it would be great.

Also, from your results I can see that the number of FP is very high. How can this be improved? Through the training on a larger dataset? Thanks a lot in advance Giorgio

Il giorno gio 19 nov 2020 alle ore 00:36 Douglas De Rizzo Meneghetti < notifications@github.com> ha scritto:

Hello there. I just re-ran one experiment with a YOLOv3 model. The results of the command are below. I had some trouble downloading the weights file from this repo, because I used Git LFS to upload the files (they are very large) and apparently I have "exceeded my storage quota" or something, so I suggest you take a look if your weights files are all intact.

If not, I can provide you with a download link so you can test them with the weights file I used in this example (should be the same).

~/Code/darknet/darknet detector map JINT2020-ball-detection/data/yolo/obj.data JINT2020-ball-detection/networks/yolo/configs/yolov3.cfg JINT2020-ball-detection/networks/yolo/weights_trained/yolov3_final.weights -points 0 GPU isn't used Used AVX Used FMA & AVX2 OpenCV version: 4.5.0 mini_batch = 1, batch = 1, time_steps = 1, train = 0 layer filters size/strd(dil) input output 0 conv 32 3 x 3/ 1 416 x 416 x 3 -> 416 x 416 x 32 0.299 BF 1 conv 64 3 x 3/ 2 416 x 416 x 32 -> 208 x 208 x 64 1.595 BF 2 conv 32 1 x 1/ 1 208 x 208 x 64 -> 208 x 208 x 32 0.177 BF 3 conv 64 3 x 3/ 1 208 x 208 x 32 -> 208 x 208 x 64 1.595 BF 4 Shortcut Layer: 1, wt = 0, wn = 0, outputs: 208 x 208 x 64 0.003 BF 5 conv 128 3 x 3/ 2 208 x 208 x 64 -> 104 x 104 x 128 1.595 BF 6 conv 64 1 x 1/ 1 104 x 104 x 128 -> 104 x 104 x 64 0.177 BF 7 conv 128 3 x 3/ 1 104 x 104 x 64 -> 104 x 104 x 128 1.595 BF 8 Shortcut Layer: 5, wt = 0, wn = 0, outputs: 104 x 104 x 128 0.001 BF 9 conv 64 1 x 1/ 1 104 x 104 x 128 -> 104 x 104 x 64 0.177 BF 10 conv 128 3 x 3/ 1 104 x 104 x 64 -> 104 x 104 x 128 1.595 BF 11 Shortcut Layer: 8, wt = 0, wn = 0, outputs: 104 x 104 x 128 0.001 BF 12 conv 256 3 x 3/ 2 104 x 104 x 128 -> 52 x 52 x 256 1.595 BF 13 conv 128 1 x 1/ 1 52 x 52 x 256 -> 52 x 52 x 128 0.177 BF 14 conv 256 3 x 3/ 1 52 x 52 x 128 -> 52 x 52 x 256 1.595 BF 15 Shortcut Layer: 12, wt = 0, wn = 0, outputs: 52 x 52 x 256 0.001 BF 16 conv 128 1 x 1/ 1 52 x 52 x 256 -> 52 x 52 x 128 0.177 BF 17 conv 256 3 x 3/ 1 52 x 52 x 128 -> 52 x 52 x 256 1.595 BF 18 Shortcut Layer: 15, wt = 0, wn = 0, outputs: 52 x 52 x 256 0.001 BF 19 conv 128 1 x 1/ 1 52 x 52 x 256 -> 52 x 52 x 128 0.177 BF 20 conv 256 3 x 3/ 1 52 x 52 x 128 -> 52 x 52 x 256 1.595 BF 21 Shortcut Layer: 18, wt = 0, wn = 0, outputs: 52 x 52 x 256 0.001 BF 22 conv 128 1 x 1/ 1 52 x 52 x 256 -> 52 x 52 x 128 0.177 BF 23 conv 256 3 x 3/ 1 52 x 52 x 128 -> 52 x 52 x 256 1.595 BF 24 Shortcut Layer: 21, wt = 0, wn = 0, outputs: 52 x 52 x 256 0.001 BF 25 conv 128 1 x 1/ 1 52 x 52 x 256 -> 52 x 52 x 128 0.177 BF 26 conv 256 3 x 3/ 1 52 x 52 x 128 -> 52 x 52 x 256 1.595 BF 27 Shortcut Layer: 24, wt = 0, wn = 0, outputs: 52 x 52 x 256 0.001 BF 28 conv 128 1 x 1/ 1 52 x 52 x 256 -> 52 x 52 x 128 0.177 BF 29 conv 256 3 x 3/ 1 52 x 52 x 128 -> 52 x 52 x 256 1.595 BF 30 Shortcut Layer: 27, wt = 0, wn = 0, outputs: 52 x 52 x 256 0.001 BF 31 conv 128 1 x 1/ 1 52 x 52 x 256 -> 52 x 52 x 128 0.177 BF 32 conv 256 3 x 3/ 1 52 x 52 x 128 -> 52 x 52 x 256 1.595 BF 33 Shortcut Layer: 30, wt = 0, wn = 0, outputs: 52 x 52 x 256 0.001 BF 34 conv 128 1 x 1/ 1 52 x 52 x 256 -> 52 x 52 x 128 0.177 BF 35 conv 256 3 x 3/ 1 52 x 52 x 128 -> 52 x 52 x 256 1.595 BF 36 Shortcut Layer: 33, wt = 0, wn = 0, outputs: 52 x 52 x 256 0.001 BF 37 conv 512 3 x 3/ 2 52 x 52 x 256 -> 26 x 26 x 512 1.595 BF 38 conv 256 1 x 1/ 1 26 x 26 x 512 -> 26 x 26 x 256 0.177 BF 39 conv 512 3 x 3/ 1 26 x 26 x 256 -> 26 x 26 x 512 1.595 BF 40 Shortcut Layer: 37, wt = 0, wn = 0, outputs: 26 x 26 x 512 0.000 BF 41 conv 256 1 x 1/ 1 26 x 26 x 512 -> 26 x 26 x 256 0.177 BF 42 conv 512 3 x 3/ 1 26 x 26 x 256 -> 26 x 26 x 512 1.595 BF 43 Shortcut Layer: 40, wt = 0, wn = 0, outputs: 26 x 26 x 512 0.000 BF 44 conv 256 1 x 1/ 1 26 x 26 x 512 -> 26 x 26 x 256 0.177 BF 45 conv 512 3 x 3/ 1 26 x 26 x 256 -> 26 x 26 x 512 1.595 BF 46 Shortcut Layer: 43, wt = 0, wn = 0, outputs: 26 x 26 x 512 0.000 BF 47 conv 256 1 x 1/ 1 26 x 26 x 512 -> 26 x 26 x 256 0.177 BF 48 conv 512 3 x 3/ 1 26 x 26 x 256 -> 26 x 26 x 512 1.595 BF 49 Shortcut Layer: 46, wt = 0, wn = 0, outputs: 26 x 26 x 512 0.000 BF 50 conv 256 1 x 1/ 1 26 x 26 x 512 -> 26 x 26 x 256 0.177 BF 51 conv 512 3 x 3/ 1 26 x 26 x 256 -> 26 x 26 x 512 1.595 BF 52 Shortcut Layer: 49, wt = 0, wn = 0, outputs: 26 x 26 x 512 0.000 BF 53 conv 256 1 x 1/ 1 26 x 26 x 512 -> 26 x 26 x 256 0.177 BF 54 conv 512 3 x 3/ 1 26 x 26 x 256 -> 26 x 26 x 512 1.595 BF 55 Shortcut Layer: 52, wt = 0, wn = 0, outputs: 26 x 26 x 512 0.000 BF 56 conv 256 1 x 1/ 1 26 x 26 x 512 -> 26 x 26 x 256 0.177 BF 57 conv 512 3 x 3/ 1 26 x 26 x 256 -> 26 x 26 x 512 1.595 BF 58 Shortcut Layer: 55, wt = 0, wn = 0, outputs: 26 x 26 x 512 0.000 BF 59 conv 256 1 x 1/ 1 26 x 26 x 512 -> 26 x 26 x 256 0.177 BF 60 conv 512 3 x 3/ 1 26 x 26 x 256 -> 26 x 26 x 512 1.595 BF 61 Shortcut Layer: 58, wt = 0, wn = 0, outputs: 26 x 26 x 512 0.000 BF 62 conv 1024 3 x 3/ 2 26 x 26 x 512 -> 13 x 13 x1024 1.595 BF 63 conv 512 1 x 1/ 1 13 x 13 x1024 -> 13 x 13 x 512 0.177 BF 64 conv 1024 3 x 3/ 1 13 x 13 x 512 -> 13 x 13 x1024 1.595 BF 65 Shortcut Layer: 62, wt = 0, wn = 0, outputs: 13 x 13 x1024 0.000 BF 66 conv 512 1 x 1/ 1 13 x 13 x1024 -> 13 x 13 x 512 0.177 BF 67 conv 1024 3 x 3/ 1 13 x 13 x 512 -> 13 x 13 x1024 1.595 BF 68 Shortcut Layer: 65, wt = 0, wn = 0, outputs: 13 x 13 x1024 0.000 BF 69 conv 512 1 x 1/ 1 13 x 13 x1024 -> 13 x 13 x 512 0.177 BF 70 conv 1024 3 x 3/ 1 13 x 13 x 512 -> 13 x 13 x1024 1.595 BF 71 Shortcut Layer: 68, wt = 0, wn = 0, outputs: 13 x 13 x1024 0.000 BF 72 conv 512 1 x 1/ 1 13 x 13 x1024 -> 13 x 13 x 512 0.177 BF 73 conv 1024 3 x 3/ 1 13 x 13 x 512 -> 13 x 13 x1024 1.595 BF 74 Shortcut Layer: 71, wt = 0, wn = 0, outputs: 13 x 13 x1024 0.000 BF 75 conv 512 1 x 1/ 1 13 x 13 x1024 -> 13 x 13 x 512 0.177 BF 76 conv 1024 3 x 3/ 1 13 x 13 x 512 -> 13 x 13 x1024 1.595 BF 77 conv 512 1 x 1/ 1 13 x 13 x1024 -> 13 x 13 x 512 0.177 BF 78 conv 1024 3 x 3/ 1 13 x 13 x 512 -> 13 x 13 x1024 1.595 BF 79 conv 512 1 x 1/ 1 13 x 13 x1024 -> 13 x 13 x 512 0.177 BF 80 conv 1024 3 x 3/ 1 13 x 13 x 512 -> 13 x 13 x1024 1.595 BF 81 conv 18 1 x 1/ 1 13 x 13 x1024 -> 13 x 13 x 18 0.006 BF 82 yolo [yolo] params: iou loss: mse (2), iou_norm: 0.75, obj_norm: 1.00, cls_norm: 1.00, delta_norm: 1.00, scale_x_y: 1.00 83 route 79 -> 13 x 13 x 512 84 conv 256 1 x 1/ 1 13 x 13 x 512 -> 13 x 13 x 256 0.044 BF 85 upsample 2x 13 x 13 x 256 -> 26 x 26 x 256 86 route 85 61 -> 26 x 26 x 768 87 conv 256 1 x 1/ 1 26 x 26 x 768 -> 26 x 26 x 256 0.266 BF 88 conv 512 3 x 3/ 1 26 x 26 x 256 -> 26 x 26 x 512 1.595 BF 89 conv 256 1 x 1/ 1 26 x 26 x 512 -> 26 x 26 x 256 0.177 BF 90 conv 512 3 x 3/ 1 26 x 26 x 256 -> 26 x 26 x 512 1.595 BF 91 conv 256 1 x 1/ 1 26 x 26 x 512 -> 26 x 26 x 256 0.177 BF 92 conv 512 3 x 3/ 1 26 x 26 x 256 -> 26 x 26 x 512 1.595 BF 93 conv 18 1 x 1/ 1 26 x 26 x 512 -> 26 x 26 x 18 0.012 BF 94 yolo [yolo] params: iou loss: mse (2), iou_norm: 0.75, obj_norm: 1.00, cls_norm: 1.00, delta_norm: 1.00, scale_x_y: 1.00 95 route 91 -> 26 x 26 x 256 96 conv 128 1 x 1/ 1 26 x 26 x 256 -> 26 x 26 x 128 0.044 BF 97 upsample 2x 26 x 26 x 128 -> 52 x 52 x 128 98 route 97 36 -> 52 x 52 x 384 99 conv 128 1 x 1/ 1 52 x 52 x 384 -> 52 x 52 x 128 0.266 BF 100 conv 256 3 x 3/ 1 52 x 52 x 128 -> 52 x 52 x 256 1.595 BF 101 conv 128 1 x 1/ 1 52 x 52 x 256 -> 52 x 52 x 128 0.177 BF 102 conv 256 3 x 3/ 1 52 x 52 x 128 -> 52 x 52 x 256 1.595 BF 103 conv 128 1 x 1/ 1 52 x 52 x 256 -> 52 x 52 x 128 0.177 BF 104 conv 256 3 x 3/ 1 52 x 52 x 128 -> 52 x 52 x 256 1.595 BF 105 conv 18 1 x 1/ 1 52 x 52 x 256 -> 52 x 52 x 18 0.025 BF 106 yolo [yolo] params: iou loss: mse (2), iou_norm: 0.75, obj_norm: 1.00, cls_norm: 1.00, delta_norm: 1.00, scale_x_y: 1.00 Total BFLOPS 65.304 avg_outputs = 516723 Loading weights from JINT2020-ball-detection/networks/yolo/weights_trained/yolov3_final.weights... seen 64, trained: 384 K-images (6 Kilo-batches_64) Done! Loaded 107 layers from weights-file

calculation mAP (mean average precision)... Detection layer: 82 - type = 28 Detection layer: 94 - type = 28 Detection layer: 106 - type = 28 252 detections_count = 414, unique_truth_count = 250 class_id = 0, name = ball, ap = 13.95% (TP = 49, FP = 207)

for conf_thresh = 0.25, precision = 0.19, recall = 0.20, F1-score = 0.19 for conf_thresh = 0.25, TP = 49, FP = 207, FN = 201, average IoU = 14.22 %

IoU threshold = 50 %, used Area-Under-Curve for each unique Recall mean average precision (mAP@0.50) = 0.139467, or 13.95 % Total Detection Time: 191 Seconds

Set -points flag: -points 101 for MS COCO -points 11 for PascalVOC 2007 (uncomment difficult in voc.data) -points 0 (AUC) for ImageNet, PascalVOC 2010-2012, your custom dataset

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/douglasrizzo/JINT2020-ball-detection/issues/1#issuecomment-730046256, or unsubscribe https://github.com/notifications/unsubscribe-auth/AHGJZDB3WKV52HWD7DYGNRLSQRSAXANCNFSM4T2MRONQ .

douglasrizzo commented 3 years ago

Try the following link. I'm not very familiar with Microsoft OneDrive/Sharepoint, I hope it works: https://feiedu-my.sharepoint.com/:f:/g/personal/douglasrizzo_fei_edu_br/EoybKMxzPzNAlO_Umo85pPQBDlm7Oxht_cFwo4IqVnSKZQ?e=oBNadC

As for making the network even better, I think it's possible to achieve considerably higher mAP values using just the the provided data set. I was able to do so with the MobileNets, achieving as high as 0.88 with MobileNetV3. However, in the YOLO models, I constantly got lower mAP values, like the 0.139467 you see in the output above.

When preparing the YOLO models for training, I followed AlexeyAB's tutorial to the letter. One thing I found a little strange is that, given my data set size and number of classes, the README told me to set the max_batches parameter to 6000, which seems kinda low. If you're willing to train the model using the provided data set, I suggest you try ignoring the README and train for a longer period of time, maybe for a whole day or two on a decent GPU.

Having a larger data set is also nice, but if that's unfeasible, I suggest you look for image augmentation techniques, such as imgaug or this repo.

giorgiomondauto commented 3 years ago

Hey, great thanks I was able to download the weights and replicate your results. For the training, I may following your suggestions. Before doing that though, I would like to try now to do the same with MobileNetv3 as you suggested.

I noticed that below python ~/detection_util_scripts/generate_pbtxt.py csv JINT2020-ball-detection/data/train.csv JINT2020-ball-detection/data/**mobilenets/**data.pbtxt should be: python ~/detection_util_scripts/generate_pbtxt.py csv JINT2020-ball-detection/data/train.csv JINT2020-ball-detection/data/data.pbtxt

I am not really familiar with MobileNetV3. I was following your steps to run it but I came across a few challenges. Did you fine-tune MobileNetV3 on your dataset as well and got new weights right? Are the steps listed in the repo all the steps to successfully run MobileNet?

Thanks Giorgio

douglasrizzo commented 3 years ago

Did you fine-tune MobileNetV3 on your dataset as well and got new weights right?

Yes, I got a model that was pre-trained on the COCO data set and trained it on the soccer ball data set.

Are the steps listed in the repo all the steps to successfully run MobileNet?

Not necessarily, I just roughly listed the steps necessary to reproduce the results of the paper this repo is related to. This repo is not intended to provide a fully fledged soccer ball detection system.

If you're interested in running a MobileNet with the weight files I provided here, I suggest you familiarize yourself with the TensorFlow Object Detection API. You can also consider using this package I created, which makes running neural networks from the TensorFlow Object Detection API a little easier.