Questions tagged [retinanet]

41 questions
5
votes
1 answer

How to run Keras Retinanet object detection code example with custom dataset?

I'm trying to implement the Keras Retinanet example from the source: https://keras.io/examples/vision/retinanet/ and I do get the example run perfectly with the COCO-dataset. Now I would like to run the example with my own custom object detection…
jjepsuomi
  • 4,223
  • 8
  • 46
  • 74
3
votes
0 answers

Unable to find docker image

I am new to docker and learning about it recently. When I was trying to use it for one of my assignments. I got the following error of which I did not find any good solution anywhere. I have properly installed nvidia-docker and build the image…
Saurav Rai
  • 2,171
  • 1
  • 15
  • 29
2
votes
1 answer

TorchVision using pretrained weights for entire model vs backbone

TorchVision Detection models have a weights and a weights_backbone parameter. Does using pretrained weights imply that the model uses pretrained weights_backbone under the hood? I am training a RetinaNet model and um unsure which of the two options…
Harry Stuart
  • 1,781
  • 2
  • 24
  • 39
2
votes
0 answers

Torchvision RetinaNet predicts unwanted class background

I want to train the pretrained RetinaNet from torchvision with my custom dataset with 2 classes (without background). To train with RetinaNet, I did follow modifications: num_classes = 3 # num of objects to identify + background class model =…
2
votes
1 answer

Combining two different deep learning models for evaluation

I have 2 different Deep Learning Model for defect detection .One is the Object detection model and other one is Semantic segmentation model. The aim is to integrate both into single prediction algorithm. I am looking for the best way to combine…
Mansi
  • 21
  • 2
2
votes
4 answers

Tensorflow 2.1 Failed to get convolution algorithm. This is probably because cuDNN failed to initialize

I am using anaconda python 3.7 and tensorflow 2.1 with cuda 10.1 and cudnn 7.6.5, and trying to run the retinaset (https://github.com/fizyr/keras-retinanet): python keras_retinanet/bin/train.py --freeze-backbone --random-transform --batch-size 8…
jwm
  • 4,832
  • 10
  • 46
  • 78
2
votes
1 answer

COWC Dataset annotation

I'm new to deep learning. Currently, I am doing a project to detect cars in aerial imagery using the Retinanet model for that I have planned to use COWC Dataset. I have doubt in the annotation part, for now I am using labelImg annotation tool to…
2
votes
0 answers

Convert Python List to CSV

I want to train my own dataset using retinanet. The Retinanet github page says that I need to convert my dataset's annotations to csv format. Retinanet csv format: path/to/image.jpg,x1,y1,x2,y2,class_name I've converted my own dataset's…
1
vote
0 answers

TF-lite model fails to invoke with RuntimeError

I am training a detection model using the keras-retinanet library. The results are promising and now I intend to convert the TensorFlow model to TensorFlow Lite for deployment. After the conversion, the model's input details are as…
davipeix
  • 55
  • 7
1
vote
1 answer

AttributeError: 'DataParallel' object has no attribute 'copy'

I am trying to resume training monkAI pytorch retinanet. I have loaded with .pt file instead of actual model. The changes are made in Monk_Object_Detection/5_pytorch_retinanet/lib/train_detector.py, check for '# change' in the places where its…
van
  • 15
  • 9
1
vote
0 answers

What exactly is SSD_ResNet50_v1_FPN?

In the TensorFlow Models Zoo, the object detection has a few popular single shot object detection models named "retinanet/resnet50_v1_fpn_ ..." or "Retinanet (SSD with Resnet 50 v1)". The paper usually linked to these works is here but the paper…
1
vote
0 answers

How to make transferlearning with resnet50_coco_best_v2.1.0

I've being messing around on how to make transferleaning from a saved model. My case is resnet50_coco_best_v2.1.0.h5, aftear loading it like and then I try to add it to my new model resnet50Coco = models.load_model(url_model,…
1
vote
0 answers

What should I do when I'm getting an mAP of 0.000 using keras-retinanet / resnet50?

I am currently using fizyr/retinanet to train a model that detects 3 classes. When I train the model, I receive precisions of 0.0000 on all my classes. In some rounds of training, I received slightly higher precisions e.g. 0.0007. I have looked at…
AndyBB
  • 11
  • 1
1
vote
1 answer

How are the FCN heads convolved over RetinaNet's FPN features?

I've recently read the RetinaNet paper and I have yet to understood one minor detail: We have the multi-scale feature maps obtained from the FPN (P2,...P7). Then the two FCN heads (the classifier head and regessor head) are convolving each one of…
Jjang
  • 11,250
  • 11
  • 51
  • 87
1
vote
1 answer

How to install or make working keras_retinanet.utils.gpu module?

I am trying to train M-RCNN using Keras and RetinaNet as in the Keras founder's example. I installed retinanet package but this module isn't found. May be it is related to that my TF doesn't use GPU? How to check that? from keras_retinanet.utils.gpu…
1
2 3