Ask Your Question

dkurt's profile - activity

2019-10-15 04:20:10 -0500 commented question OpenVINO IR support

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

2019-09-03 05:03:33 -0500 commented question OpenVino inference engine fails on Raspberry Pi 4+NCS 2+OpenCV compiled from source

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

2019-08-20 07:09:14 -0500 commented question Cannot use Tensorflow model with batch normalization

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

2019-07-27 05:23:21 -0500 answered a question Access output of a Caffe Slice Layer

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

2019-07-19 23:49:52 -0500 commented answer Can't create layer "leaky_0/LeakyRelu" of type "LeakyRelu" in function 'getLayerInstance'

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

2019-07-04 22:49:25 -0500 commented question Error while running converted .h5 to .pb graph with OpenCV's dnn module

Please share the model if it's possible.

2019-06-24 22:40:35 -0500 commented question Dnn tiny yolo selecting between graylevel and color image c++

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

2019-06-09 03:01:11 -0500 commented question Are All the Onnx 1.5 implemented operations supported by the OpenCv 4.1 readNetfromOnnx Importer?

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

2019-06-08 15:46:05 -0500 commented question Are All the Onnx 1.5 implemented operations supported by the OpenCv 4.1 readNetfromOnnx Importer?

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

2019-05-22 07:36:58 -0500 commented question TinyYolo Tiny gives long process time once in a while

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

2019-05-22 03:36:46 -0500 commented answer how to reshape a mat like (1,3,height,width) to (height,width,3)?

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

2019-05-21 13:01:22 -0500 answered a question Tensorflow lite Graph with OpenCV DNN

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

2019-05-12 03:24:56 -0500 commented question Error while importing network from Caffe

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

2019-05-08 23:01:04 -0500 commented answer what is the change of dnn module in opencv4.1?

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

2019-05-08 22:59:18 -0500 answered a question how to reshape a mat like (1,3,height,width) to (height,width,3)?

Try cv::dnn::imagesFromBlob

2019-05-07 23:15:11 -0500 commented answer what is the change of dnn module in opencv4.1?

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

2019-05-07 08:39:14 -0500 commented answer readNet fails

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

2019-05-06 06:37:50 -0500 answered a question readNet fails

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

2019-05-06 00:36:41 -0500 answered a question How to process output of detection network when batch of images is used as network input

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

2019-04-24 10:09:57 -0500 commented question net.forward() crash in Faster-RCNN Object Detection Sample

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

2019-04-23 04:40:46 -0500 answered a question Where can I find the cmake options used for the 4.1.0 build

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

2019-04-22 03:10:34 -0500 commented answer blobFromImage is giving wrong value.

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

2019-04-22 02:15:57 -0500 commented answer blobFromImage is giving wrong value.

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

2019-04-22 02:10:51 -0500 commented question How do I read .npy file using OpenCv?

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

2019-04-22 02:10:12 -0500 commented question How do I read .npy file using OpenCv?

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

2019-04-19 01:01:51 -0500 commented answer what is the change of dnn module in opencv4.1?

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

2019-04-19 01:00:15 -0500 commented answer what is the change of dnn module in opencv4.1?

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

2019-04-19 00:53:38 -0500 edited question what is the change of dnn module in opencv4.1?

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

2019-04-17 06:27:40 -0500 commented answer what is the change of dnn module in opencv4.1?

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

2019-04-17 01:30:13 -0500 edited answer what is the change of dnn module in opencv4.1?

with dilated convolutions So the models are faster_rcnn_inception_resnet_v2_atrous_coco mask_rcnn_resnet101_atro

2019-04-17 01:29:42 -0500 answered a question what is the change of dnn module in opencv4.1?

with dilated convolutions So the models are faster_rcnn_inception_resnet_v2_atrous_coco mask_rcnn_resnet101_atrou

2019-04-15 09:41:11 -0500 commented answer Faster RCNN no results

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

2019-04-15 09:40:30 -0500 commented answer Faster RCNN no results

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

2019-04-15 04:27:05 -0500 answered a question Faster RCNN no results

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

2019-04-12 12:29:41 -0500 commented question Faster RCNN no results

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

2019-04-11 09:05:49 -0500 answered a question Where can I find how to deal with the return of the forward method?

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

2019-04-10 08:41:40 -0500 commented question error: (-215:Assertion failed) (int)_numAxes == inp0.dims in function 'cv::dnn::PermuteLayerImpl::finalize'

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

2019-04-09 09:28:08 -0500 answered a question layer tf.shape forbidden in opencv?

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

2019-04-09 09:04:12 -0500 commented question layer tf.shape forbidden in opencv?

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

2019-04-09 08:58:05 -0500 commented question Build error with Opencv with Inference Engine backend on raspberry

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

2019-04-09 08:57:35 -0500 commented question Build error with Opencv with Inference Engine backend on raspberry

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

2019-04-08 10:06:09 -0500 commented question Build error with Opencv with Inference Engine backend on raspberry

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

2019-04-08 10:05:36 -0500 commented question Build error with Opencv with Inference Engine backend on raspberry

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

2019-04-08 04:18:32 -0500 commented question Build error with Opencv with Inference Engine backend on raspberry

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

2019-04-05 13:03:06 -0500 commented answer QR code detection in Java (Android Studio)

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

2019-04-05 05:49:26 -0500 commented answer QR code detection in Java (Android Studio)

@berak, @Agus, Proposed a hot fix here: https://github.com/opencv/opencv/pull/14266. (imho, in the long run, the jav

2019-04-05 04:47:29 -0500 commented answer QR code detection in Java (Android Studio)

Oh, that's looks like a lack of tests because in the master branch I got: // // C++: string cv::QRCodeDetector::decode

2019-04-05 04:42:37 -0500 commented answer QR code detection in Java (Android Studio)

@berak, For 3.4 branch I have: // // C++: String cv::QRCodeDetector::decode(Mat img, Mat points, Mat& straight_qrc

2019-04-04 08:52:27 -0500 answered a question Can't create layer "leaky_0/LeakyRelu" of type "LeakyRelu" in function 'getLayerInstance'

Yeah, looks like we need to modify tf_importer.cpp. On the other hand you can try define LeakyRelu layer to ReLU impleme

2019-04-04 03:33:52 -0500 answered a question DNN with tiny yolo 3

YOLOv3 supports starts from OpenCV 3.4.2: https://github.com/opencv/opencv/wiki/ChangeLog#version342