kentaroy47/frcnn-from-scratch-with-keras

Issues

• Can't train RPN separately (train_rpn.py)

  #108 opened 9 months ago
  0

• SystemExit: 2

  #107 opened 2 years ago
  0

• AttributeError: module 'tensorflow' has no attribute 'ConfigProto', how can I fix it ?

  #106 opened 2 years ago
  1

• Unable to reproduce performance with PASCALVOC2010 and backbone resnet

  #105 opened 3 years ago
  0

• Anchor box generation when calculating the rpn

  #104 opened 3 years ago
  1

• RPN is not producing bounding boxes that overlap the ground truth boxes. Check RPN settings or keep training.

  #103 opened 3 years ago
  0

• Error during testing

  #102 opened 3 years ago
  0

• Problem with AttributeError: 'str' object has no attribute 'shape'

  #101 opened 3 years ago
  2

• Why is the epoch_length is choosen as 1000? when should we change this?

  #100 opened 3 years ago
  1

• ModuleNotFoundError: No module named 'tensorflow.compat.v1'

  #99 opened 4 years ago
  1

• Modify Network for Multi-task Learning

  #98 opened 4 years ago
  0

• Fit automated method instead of manually write it.

  #97 opened 4 years ago
  0

• Cross Validation

  #96 opened 4 years ago
  1

• How to plot feature maps from RPN model?

  #95 opened 4 years ago
  1

• Write the values of the predicted bounding boxes in measure_map

  #94 opened 4 years ago
  0

• Keyerror using measure_map for custom data using the trained VGG16 model

  #93 opened 4 years ago
  0

• Attempting to fetch value instead of handling error Internal: no supported devices found for platform CUDA

  #92 opened 4 years ago
  0

• How to write the results from the test_frcnn to a CSV file?

  #91 opened 4 years ago
  0

• RPN weights are not saved

  #90 opened 4 years ago
  2

• classifier_layers not used in resnet.py

  #89 opened 4 years ago
  2

• P_regr not used in test_frcnn

  #88 opened 4 years ago
  0

• Mistake in Loss function! ("rpn_loss_cls" function in "losses.py")

  #87 opened 4 years ago
  0

• Mistake in positive anchor box calculation (calc_rpn in data_generator.py)

  #86 opened 4 years ago
  1

• Why 4.0 for std_scaling?

  #85 opened 4 years ago
  1

• data_gen_val didn't used in "train_frcnn.py"

  #84 opened 4 years ago
  4

• ResNet-50 architecture

  #83 opened 4 years ago
  0

• how to change the receptive fields in faster rcnn?

  #82 opened 4 years ago
  1

• Custom backbone

  #81 opened 4 years ago
  1

• background class (bg) in faster rcnn

  #80 opened 4 years ago
  2

• ValueError: Layer #314 (named "time_distributed_2"), weight <tf.Variable 'time_distributed_2/kernel:0' shape=(2048, 4096) dtype=float32_ref> has shape (2048, 4096), but the saved weight has shape (4096, 4096).

  #79 opened 4 years ago
  2

• F1-score loss for RRN

  #78 opened 4 years ago
  0

• Labeling images for HD5?

  #77 opened 4 years ago
  1

• train_rpn.py resnet50 silently closed

  #76 opened 3 years ago
  4

• Set Up Keras-FasterRCNN in local machine from scratch

  #75 opened 4 years ago
  9

• Small object detection

  #74 opened 4 years ago
  0

• TensorFlow 2 and current status

  #73 opened 4 years ago
  4

• RPN, Faster R-CNN training way & ResNet-50 Arch

  #72 opened 4 years ago
  2

• measure_map.py

  #71 opened 4 years ago
  2

• CheckPoint option

  #70 opened 4 years ago
  1

• Exception: 'a' cannot be empty unless no samples are taken

  #69 opened 4 years ago
  5

• Test with images of different size

  #68 opened 4 years ago
  2

• Getting confidence score/distributions for all objects

  #67 opened 4 years ago
  2

• No detection in test_frcnn.py

  #66 opened 4 years ago
  1

• Predicted bounding boxes always squared 16x16

  #65 opened 4 years ago
  0

• timedistributed function

  #64 opened 4 years ago
  0

• loss calculation

  #63 opened 4 years ago
  1

• can you provide pre-trained model?

  #62 opened 4 years ago
  2

• Strange behavior in "rpn_to_roi" function

  #61 opened 4 years ago
  2

• Hot to convert hdf5 to pb?

  #60 opened 4 years ago
  3

• Problem with AttributeError: 'NoneType' object has no attribute 'shape'

  #59 opened 4 years ago
  4