dkurt's profile - activity

Use OpenCL cache: https://github.com/intel/compute-runtime/blob/master/opencl/doc/FAQ.md#feature-cl_cache. Generete is o

Use OpenCL cache: https://github.com/intel/compute-runtime/blob/master/opencl/doc/FAQ.md#feature-cl_cache. Generete is o

Please try to add one more method @Override protected List<? extends CameraBridgeViewBase> getCameraViewList() {

@Shay Weissman, If you want to report a bug it's better to use https://github.com/opencv/opencv/issues. But this way it

Take a look at https://github.com/opencv/opencv/pull/16184. Currently this PR in progress

It supports. Please check https://github.com/opencv/opencv/wiki/Intel%27s-Deep-Learning-Inference-Engine-backend.

Please try the sample first https://github.com/opencv/opencv/blob/master/samples/dnn/object_detection.py

Feel free to open an issue providing steps to reproduce it (attach the model). We observed several times buggy Keras bat

Try to use forwardAndRetrieve: https://docs.opencv.org/master/db/d30/classcv_1_1dnn_1_1Net.html#a00e707a86b2da4f980f9342

Feel free to open an issue on GitHub. Please add a model there as well.

Please share the model if it's possible.

Maybe you can try to get the first convolution layer and determine input channels number by it's weights.

Do you already have a network? It's better to start with specific model to say if it's supported or not yet. Anyway ther

Please provide us at least the list of all the ONNX operations mentioned in the topic.

Please provide some detailed report about efficiency. At least the inference time measurements.

@lunasdejavu, Cannot reproduce. Here is my script: import cv2 as cv import numpy as np batch_size = 16 blob = np.rando

Currently import of TFLite models is not supported. You may track corresponding feature request: https://github.com/open

Start with OpenCV sample https://github.com/opencv/opencv/blob/master/samples/dnn/object_detection.py. Faster takes two

Please use a documentation: https://docs.opencv.org/master/d6/d0f/group__dnn.html#ga65a1da76cb7d6852bdf7abbd96f19084

Try cv::dnn::imagesFromBlob

PyTorch model should be converted to ONNX. Use cv::dnn::readNet or readNetFromONNX

May I ask you to propose a PR which extends regex matching for this group of layers? Thanks!

It looks like a model created with TensorFlow Object Detection API. Do you have a .config file? See https://github.com/o

There is no cols and rows for cv::Mat with number of dimensions >2 (source). Check blob.size[0], blob.size[1] and blo

Hmm, it's a bit strange because OpenVINO 2019R1 has a bug which don't let me run Faster-RCNN network with the following

There is builbot where you can find all the details: https://pullrequest.opencv.org/buildbot/waterfall?tag=pack-master

Where is my favorite :confused: emojis from GitHub when it's so actual :)

@my_opencv, From the docs: the number of rows and columns or (-1, -1) when the matrix has more than 2 dimensions sour

@Jozay, there are some rules and guidelines of how to use the forums. Nobody want to do your work: Whenever you are in

@Jozay, there are some rules and guidelines of how to use the forums. Nobody want to do your work: Whenever you are in

@lunasdejavu, feel free to open a feature request at https://github.com/opencv/opencv/issues. It might be a complicated

@lunasdejavu, feel free to open a feature request at https://github.com/opencv/opencv/issues. It might be a complicated

what is the change of dnn module in opencv4.1? the changlog of opencv 4.1 : New networks from TensorFlow Object Detectio

@lunasdejavu, It's a text version of binary .pb files. They are serialized with protobuf. I think you can use existing .

with dilated convolutions So the models are faster_rcnn_inception_resnet_v2_atrous_coco mask_rcnn_resnet101_atro

with dilated convolutions So the models are faster_rcnn_inception_resnet_v2_atrous_coco mask_rcnn_resnet101_atrou

It has. The sample just prints confidences without class names. Try to pass a file object_detection_classes_coco.txt.

If has. The sample just prints confidences without class names. Try to pass a file object_detection_classes_coco.txt.

@tobix10, Thanks for pointing that issue! The https://github.com/opencv/opencv/blob/3.4.6/samples/dnn/object_detection.

How we can reproduce it? Is it a public model? Can you provide at least an image sample?

There are code examples using SDD, Mask-RCNN and YOLOv3 on the internet So that is why there are samples in the Ope

Can you provide a reproducer? In example, upload a .pb file?

May I ask you to try to replace reshape = tf.reshape(groupement2,[tailleLots, tailleGroupement2[1] * tailleGroupement2[

In the regular cases Shape node is used in subgraphs. In example, Flatten operation in TensorFlow might looks like a sub

@potoo, Unfortunately I can say nothing without reproducer. Let's create a separate thread if you don't mind.

@potoo, Unfortunately can say nothing without reproducer. Let's create a separate thread if you doesn't mind.

@potoo, Have you followed https://github.com/opencv/opencv/wiki/Intel%27s-Deep-Learning-Inference-Engine-backend#raspbia

@potoo, Have you followed https://github.com/opencv/opencv/wiki/Intel%27s-Deep-Learning-Inference-Engine-backend#raspbia

@potoo, it looks like a conflict with IE versions. Please show which you downloaded and which is detected by OpenCV (see

Check a test code from the PR. Method detectAndDecode returns a string from an image with QRCode.