Answered
How can I merge (concatenate) feature maps from two different CNN layers in my architecture?
Hi, In R2017b, the depth concatenation layer can be used to merge feature maps: https://www.mathworks.com/help/nnet/ref/n...

6 years ago | 1

| accepted

Answered
Is there any example about fcnLayers and its predictions for semantic segmentation
In R2017b, we cannot train with arbitrary size images. This is a feature we will consider for a future release. We will also loo...

6 years ago | 0

Answered
Mini-batch loss not changing for FCN
This is a R2017b bug in fcnLayers. I've attached a patch for this issue. Here are the instructions for patching R2017b: 1...

6 years ago | 0

Answered
Invalid entry in label data table for Scene labelType when using Image Labeler app?
This is a bug in R2017b. The attached patch, scene_label_patch_17b.zip, patches this bug in R2017b. Apply the patch using th...

6 years ago | 1

| accepted

Answered
Why ocr function doesn't recognize the numbers?
Hi, A little bit of pre-processing and using ROIs to specify where the words are will help. By default, OCR uses page layout...

6 years ago | 4

| accepted

Answered
Error using trainFastRCNNObjectDetector
It's likely that your custom region proposal method is producing ROIs that are too small to process. The minimum size that we ca...

6 years ago | 2

| accepted

Answered
How to implement Faster R CNN object detector?
Hi Jay, Because you have objects that have a single size, 96x96, you can set BoxPyramidScale to 1 and NumBoxPyramidLevels to ...

7 years ago | 0

Answered
How to plot Hog features in the image
Hi, The following example, will visualize the HOG features at a set of points: I2 = imread('gantrycrane.png'); ...

7 years ago | 0

Answered
multiple gpu slow?
Hi, This is due to a limitation with NVIDIA’s GPU-to-GPU communication on Windows. If you have the option, you should consid...

7 years ago | 1

Answered
ocrTrainer 'Not enough input arguments'
Hi Michal, I can look into what's causing the error. Are you able to share your training data and training session file (save...

7 years ago | 0

Answered
Why won't vision.DeployableVideoPlayer work in MATLAB compiler deployed executable?
Hi, That's strange. What happens if you try to reduce the image size from 2048x2048x3 to something like 10x10x3? Does it stil...

8 years ago | 0

Answered
minimum background ratio in foreground detection
Hi, It'll be more helpful to answer your second question first. To start, see here for the definition of a multimodal distrib...

8 years ago | 1

Answered
Unable to run bag of features
Hi Leonard, The error occurs because bagOfFeatures is extracting features from all of your training images and storing them i...

8 years ago | 1

| accepted

Answered
Problem with evaluate(imagecategoryclassifier) function from Vision System tool box
Hi Jacob, Thanks for reporting this issue. This is a bug. The predicted labels are correct, but the scores are not. I can...

8 years ago | 1

| accepted

Answered
Clarification of the detectSURFFeatures 'NumOctaves' parameter
Hi Chris, Yes, it does include octaves 1-2. The 'NumOctaves' parameter controls the total number of octaves. Hope that h...

8 years ago | 1

Answered
How to get the variance of a binaryFeature object?
Hi David, The error is due to the use of binaryFeatures in your custom extractor. bagOfFeatures does not support binaryFeatur...

9 years ago | 1

Answered
Unable to recognize letter labels in a image using ocr() function.
Hi Jack, You'll need to do a bit of pre-processing to get OCR to work on your image. Here's what I did to get it working: ...

9 years ago | 6

| accepted

Answered
Computer Vision System ToolBox 'detectMSERFeatures'
Hi, The ThresholdDelta is a percentage. For example in an 8-bit image, a value of 4 percent means the threshold increment use...

9 years ago | 1

| accepted

Answered
Foreground Detector - how do I save/load an existing background model?
Hi, You'll need to execute: feature('SystemObjectsFullSaveLoad',1); to enable the save/load to work correctly for visi...

9 years ago | 3

| accepted

Answered
How to create Matlab Simulink Model for Object Detection?
Hi Sorath, You can call the vision.CascadeObjectDetector by using a MATLAB Function block in Simulink. For example: % MATL...

9 years ago | 1

| accepted

Answered
How to automatically identify text lines in curved document images ?
Hi, You might be able to use CPSELECT and FITGEOTRANS with the polynomial transform to remove the distortion of the text line...

9 years ago | 1