Nothing found in Tensorflow Object Detection API

Question

I am trying to implement a sample Tensorflow Object Detection API. I read sentdex to get started. The sample code works fine, it also shows images that are used to test the results, but the borders around the detected objects are not displayed. Just the image of the plane is displayed without errors.

I am using this code: This is a Github link .

This is my result after running the sample code.

another image without any detection.

What am I missing here? The code is included in the link above and there are no error logs.

Results of fields, grades, classes, num in that order.

[[[ 0.74907303 0.14624023 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0.20880508 1. 1. ] [ 0.74907303 0.14624023 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0.74907303 0.14624023 1. 1. ] [ 0.74907303 0.14624023 1. 1. ] [ 0.74907303 0.14624023 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0.74907303 0.14624023 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0.74907303 0.14624023 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0.74907303 0.14624023 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0.20934391 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0.20880508 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0.74907303 0.14624023 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0.74907303 0.14624023 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0.74907303 0.14624023 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0.74907303 0.14624023 1. 1. ] [ 0.74907303 0.14624023 1. 1. ] [ 0.74907303 0.14624023 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0.74907303 0.14624023 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0.74907303 0.14624023 1. 1. ] [ 0.74907303 0.14624023 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0.74907303 0.14624023 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ]]] [[ 0.03587547 0.02224986 0.0186467 0.01096812 0.01003207 0.00654409 0.00633549 0.00534311 0.0049596 0.00410213 0.00362371 0.00339186 0.00308251 0.00303347 0.00293389 0.00277099 0.00269575 0.00266825 0.00263925 0.00263331 0.00258657 0.00240822 0.0022581 0.00186967 0.00184311 0.00180467 0.00177475 0.00173655 0.00172811 0.00171935 0.00171891 0.00170288 0.00163755 0.00162967 0.00160273 0.00156545 0.00153615 0.00140941 0.00132407 0.00131524 0.0013105 0.00129431 0.0012582 0.0012553 0.00122365 0.00119186 0.00115651 0.00115186 0.00112369 0.00107097 0.00105805 0.00104338 0.00102719 0.00102337 0.00100349 0.00097762 0.00096851 0.00092741 0.00088506 0.00087696 0.0008734 0.00084826 0.00084135 0.00083513 0.00083398 0.00082068 0.00080583 0.00078979 0.00078059 0.00077476 0.00075448 0.00074426 0.00074421 0.00070195 0.00068741 0.00068138 0.00067262 0.00067125 0.00067033 0.00066035 0.00064729 0.00064205 0.00061964 0.00061794 0.00060835 0.00060465 0.00059548 0.00059479 0.00059461 0.00059436 0.00059426 0.00059411 0.00059406 0.00059392 0.00059365 0.00059351 0.00059191 0.00058798 0.00058682 0.00058148]] [[ 1. 1. 18. 32. 62. 60. 63. 67. 61. 49. 31. 84. 50. 54. 15. 44. 44. 49. 31. 56. 88. 28. 88. 52. 17. 32. 38. 75. 3. 33. 48. 59. 35. 57. 47. 51. 19. 27. 72. 4. 84. 6. 55. 20. 58. 65. 61. 82. 42. 34. 40. 21. 43. 64. 39. 62. 36. 22. 79. 46. 16. 40. 41. 77. 16. 48. 78. 77. 89. 86. 27. 8. 87. 5. 25. 70. 80. 76. 75. 67. 65. 37. 2. 9. 73. 63. 29. 30. 69. 66. 68. 26. 71. 12. 45. 83. 13. 85. 74. 23.]] [ 100.] [[[ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0.68494415 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0.68494415 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0.00784111 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0.68494415 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0.68494415 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0.68494415 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0.68494415 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0.68494415 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0.68494415 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0. 1. 1. ] [ 0. 0.68494415 1. 1. ] [ 0. 0.68494415 1. 1. ]]] [[ 0.01044297 0.0098214 0.00942165 0.00846471 0.00613666 0.00398615 0.00357754 0.0030054 0.00255861 0.00236574 0.00232631 0.00220291 0.00185227 0.0016354 0.0015979 0.00145072 0.00143661 0.00141369 0.00122685 0.00118978 0.00108457 0.00104251 0.00099215 0.00096401 0.0008708 0.00084773 0.00080484 0.00078507 0.00078378 0.00076876 0.00072774 0.00071732 0.00071348 0.00070812 0.00069253 0.0006762 0.00067269 0.00059905 0.00059367 0.000588 0.00056114 0.0005504 0.00051472 0.00051057 0.00050973 0.00048486 0.00047297 0.00046204 0.00044787 0.00043259 0.00042987 0.00042673 0.00041978 0.00040494 0.00040087 0.00039576 0.00039059 0.00037274 0.00036831 0.00036417 0.00036119 0.00034645 0.00034479 0.00034078 0.00033771 0.00033605 0.0003333 0.0003304 0.0003294 0.00032326 0.00031787 0.00031773 0.00031748 0.00031741 0.00031732 0.00031729 0.00031724 0.00031722 0.00031717 0.00031708 0.00031702 0.00031579 0.00030416 0.00030222 0.00029739 0.00029726 0.00028289 0.0002653 0.00026325 0.00024584 0.00024221 0.00024156 0.00023911 0.00023335 0.00021619 0.0002001 0.00019127 0.00018342 0.00017273 0.00015509]] [[ 38. 1. 1. 16. 25. 38. 64. 24. 49. 56. 20. 3. 28. 2. 48. 19. 21. 62. 50. 6. 8. 7. 67. 18. 35. 53. 39. 55. 15. 57. 72. 52. 10. 5. 42. 43. 76. 22. 82. 4. 61. 23. 17. 16. 87. 62. 51. 60. 36. 58. 59. 33. 31. 54. 70. 11. 40. 79. 31. 9. 41. 77. 80. 34. 90. 89. 73. 13. 84. 32. 63. 29. 30. 69. 66. 68. 26. 71. 12. 45. 83. 14. 44. 78. 85. 46. 47. 19. 65. 74. 37. 27. 63. 88. 28. 81. 86. 75. 27. 18.]] [ 100.] 

EDIT:. According to the suggested answers, it works when we use the model faster_rcnn_resnet101_coco_2017_11_08 . But this is more accurate and why slower. I want this application with high speed, because I'm going to use it in real time (on a webcam). So I need to use a faster model ( ssd_mobilenet_v1_coco_2017_11_08 )

Answer 1

Like changing the workaround #MODEL_NAME = 'ssd_mobilenet_v1_coco_2017_11_08' to MODEL_NAME = 'faster_rcnn_resnet101_coco_2017_11_08'.

Answer 2

The problem is the model: 'ssd_mobilenet_v1_coco_2017_11_08'

Solution: change the differrent version 'ssd_mobilenet_v1_coco_11_06_2017' (this type of model is the fastest, changing to other types of models will make it slower, and not what you want)

Just change 1 line of code:

 # What model to download. MODEL_NAME = 'ssd_mobilenet_v1_coco_11_06_2017' 

When I use your code, nothing is displayed, but when I replace it with the previous experiment model 'ssd_mobilenet_v1_coco_11_06_2017' , it works fine

Answer 3

You can use the older 'ssd_mobilenet_v1 ...' and run your program completely using the boxes (I am running it now, and that is correct). This is a link to this older version. I hope that they will fix the new version soon.

Answer 4

I had the same problem.

But the new model recently uploaded it 'ssd_mobilenet_v1_coco_2017_11_17'

I tried and worked like a charm :)

Answer 5

The visualize_boxes_and_labels_on_image_array function has the following code:

  for i in range(min(max_boxes_to_draw, boxes.shape[0])): if scores is None or scores[i] > min_score_thresh: 

so the score should be greater than min_score_thresh (the default value is 0.5), you can check if there are any points more than that.