I am having trouble calling EMD() in OpenCV 2.4.2 under Mac OS ML.
I have a class with an attribute Mat _signature defined like that :
Mat _signature(size,dim+1,CV_32F);
for (int i = 0; i<size; ++i){
_signature.at<float>(i,0) = weight;
for (int j = 1; j < dim+1; ++j){
_signature.at<float>(i,j) = vec[i].at<float>(0,j-1); // vec[i] is a line vector containing the position in R^dim
}
}
I then have u and v 2 instances of that class, and when I call EMD(u._signature, v._signature, CV_DIST_L2);
It fails with OpenCV Error: One of arguments' values is out of range () in icvInitEMD, file /*SOME PATH*/OpenCV-2.4.2/modules/imgproc/src/emd.cpp, line 408
I looked at the sourcecode but could not figure out what this fails. My arguments appear in correspondence to what the documentation wants. Any help will be appreciated.
Ok, it took me quite some time to figure it out, but among my data there was a component of one of my vector which was miscalculated, and ended up being NaN.
Of course this was buried deep into my data so that it would be completely lost in any amount of data reasonably observable via a debugger (or even cout)
The cryptic error from OpenCV did the rest in confusing me.
For people stumbling upon the same issue as me :
Make sure your weight vectors are not zero
Make sure none of your data is NaN
Related
I am new to Halide and have written a simple code to compute max(127, pix(x,y)) for every pixel in an image.
Though the code runs fine on CPU, it gives me wrong outputs when I set Target::CUDA. I'm not able to find the issue.
The following is a part of my code. Let me know if there is a mistake in the code, or do I have to re-build Halide with a support that will enable CUDA.
Halide::Var x, y;
Halide::Buffer<uint8_t> inputImageBuf(inpImg, imgSizes);
Halide::Func reluOp("ReLU Operation");
reluOp(x,y) = Halide::max(127, inputImageBuf(x, y));
int numTiles = 4;
Halide::Var threads_x, threads_y, blocks_x, blocks_y;
Halide::Target targetCUDA = Halide::get_host_target();
targetCUDA.set_feature(Halide::Target::CUDA);
targetCUDA.set_feature(Halide::Target::Debug);
reluOp.gpu_tile(x, y, blocks_x, blocks_y, threads_x, threads_y, numTiles, numTiles, Halide::TailStrategy::Auto, Halide::DeviceAPI::CUDA);
// reluOp.compile_jit(targetCUDA);
reluOp.print_loop_nest();
Halide::Buffer<uint8_t> result = reluOp.realize(cols, rows, targetCUDA);
result.copy_to_host();
One thing to try is adding a inputImageBuf.set_host_dirty(). If that helps I would consider that a bug in Halide.
You can also scroll through the debug output and see if the expected number of copies to and from the host are happening.
This should be easy but I may be misunderstanding something fundamental.
I have ellipses (called movers) on a screen in box2d. The ellipse is represented by a body. The body has a variable r which I use to make the ellipse size.
I want to destroy the ellipse when the r=0.
This is what i have written:
for (int i = 0; i < movers.length; i++) {
if(movers[i].r<0){
box2d.destroyBody(movers[i].body);
}
}
However this gives me an assertion error at the line box2d.destroyBody(movers[i].body);
How do I correct this? Is this because I am destroying the body but not the ellipse possibly or some other subtlety?
So the answer is not intuituitive but for future reference the problem is that box2d gets confused when trying to kill bodies if it is done inside of a world timestep.
What does this mean?
It means that you have to add all the bodies you are using to an arrayList rather than an array and your code should work from that. If you do that then at the end of void draw you can add the following (where movers is an arrayList of objects which have bodies and r is an attribute which defines whether a body should be destroyed or not):
for (int i = movers.size()-1; i >= 0; i--) {
Mover p = movers.get(i);
if (p.r<0) { //r is whatever attribute of your body
movers.remove(i);
}
}
This will destroy the bodies by removing them from the arrayList you are working from
I'm studying Java so I'm pretty new.
I'm creating a simple 'maze' type game using GUI layouts, images, labels ect..
To create my maze layouts I used an array of strings;
mazeLayout[0] = "WWWWWWWWWW";
mazeLayout[1] = "WSSSWWSWWW";
mazeLayout[2] = "WSWSWWSSSW";
mazeLayout[3] = "WSWSWWWWSW";
mazeLayout[4] = "WSWSWWWWSW";
mazeLayout[5] = "WSWSWSSSSW";
mazeLayout[6] = "WSWSWSWWWW";
mazeLayout[7] = "WSWSWSWWWW";
mazeLayout[8] = "WSWSSSWWWW";
mazeLayout[9] = "WWWWWWWWWW";
and then converted this into a 2d array and placed a label with in image icon in it depending on the string being 'W' for wall or 'S' for space. Also the labels are an array, my thoughts behind this was for restricting movement of the player so they can't walk though walls.
int mw = 0;
int mf = 0;
for(int y = 0; y < 10; y++){
for(int x = 0; x < 10; x++){
mazeLayout2d[y][x] = mazeLayout[y].substring(x, x+1);
if (mazeLayout2d[y][x].equals("W")){
lblmazewall[mw] = new JLabel();
mazewall = new ImageIcon("mazewall.png");
lblmazewall[mw].setIcon(mazewall);
pCenter.add(lblmazewall[mw]);
mw++;
pCenter.revalidate();
}
if (mazeLayout2d[y][x].equals("S")){
lblmazefloor[mf] = new JLabel();
mazefloor = new ImageIcon("mazefloor.png");
lblmazefloor[mf].setIcon(mazefloor);
pCenter.add(lblmazefloor[mf]);
mf++;
pCenter.revalidate();
}
}
}
My problem is when i run this line
System.out.println(lblmazewall[x].getLocation()); //x being any number
I always get java.awt.Point[x=0,y=0]
I would like to know how to get the location of each wall label so i can check it against my player movement.
Is this even a valid way to do something like this?
Could someone teach me a more efficient way?
Sorry for my crude snippets and or bad programming
Thankyou Niall.
public Point getLocation()
Due to the asynchronous nature of native event handling, this method can return outdated values (for instance, after several calls of setLocation() in rapid succession). For this reason, the recommended method of obtaining a component's position is within java.awt.event.ComponentListener.componentMoved(), which is called after the operating system has finished moving the component.
The layout might not have used setLocation() internally. So that getLocation() does not return the value as expected.
I have problem about Parallel Nsight 2.2 debugger. It is very strange and I don't know how to describe it. Anyway, It works sometimes and sometimes doesn't.
What I observed is, that it works with dynamic array(this array has no effect on cuda_kernels or any other functon like cudaMemcpy atc...) named with 3 elements. And this is importnat... If I set size on 4+, it just falls down, no errors, nothing just fall down.
Interesting fact is, that if I run it normally via normal debugger hole program works correctly with right results. Also interesting fact is, that when set this array as static
unsigned topology[4];
and set in same values Nsight debugger works but very slowly.
So first of all I commented all cuda source code (like kernels and all cuda functions) but still same - it falls down. So I started to comment more host_code and I found loop (in host code) which does this creepy thing. So when program in Nsght-debug reach loop(under text) it falls down, BUT, when I write command in this loop to print number of each loop on screen, it runs, loop is finished, hole program is finished and then debugger told me:
Debug Assertion Failed!
Program:
File:f:\dd\vctools\crt_bld\self_x86\crt\src\dbgheap.c
Line: 1322
Expression: _CrtIsValidHeapPointer(pUserData)
.... I don't even have disk f ... so wtf???
Anyway, on normal debugger it runs fine and with right results.
This is mentioned loop and dynamic array *topology:
unsigned *topology;
unsigned numberOfLayersInput = 5;
topology = new unsigned [numberOfLayersInput];
topology[0] = 784;
topology[1] = 1000;
topology[2] = 800;
topology[3] = 300;
topology[4] = 10;
kernelTopology_ *topologyOfKernels;
topologyOfKernels = new kernelTopology_ [numberOfLayersInput - 1];
for (int i = 0, numberOfThreads; i < numberOfLayersInput; i++)
{
cout <<i << endl; // this is the added line!
numberOfThreads = fixedTopology[i];
topologyOfKernels[i].size = numberOfThreads;
if(numberOfThreads > THREADS_PER_BLOCK)
topologyOfKernels[i].BLOCK_SIZE = THREADS_PER_BLOCK;
else topologyOfKernels[i].BLOCK_SIZE = numberOfThreads;
if(numberOfThreads <= THREADS_PER_BLOCK)
topologyOfKernels[i].GRID_SIZE = 1;
else if(fixedTopology[i] % topologyOfKernels[i].BLOCK_SIZE == 0)
topologyOfKernels[i].GRID_SIZE = fixedTopology[i] / topologyOfKernels[i].BLOCK_SIZE;
else
topologyOfKernels[i].GRID_SIZE = (fixedTopology[i] / topologyOfKernels[i].BLOCK_SIZE) + 1;
}
I can't see any mistakes in this code... also normal debugger has no problem with it.
I have reinstalled graphics drivers, CUDA toolkit, CUDA SDK and Paralell Nsight but it does same creepy things. By the way I use Win 7 64 bit and VS2010.
Does have anyone any ideas what I should do with this?
Please, let me know if someone has any idea :)
The error
Debug Assertion Failed! Program: File:f:\dd\vctools\crt_bld\self_x86\crt\src\dbgheap.c Line: 1322
is from the Microsoft C runtime function _CrtIsValidHeapPointer. The default debug build adds additional heap and stack checks into the code. This function is used to verify that a specified pointer is in the local heap. The path f:... is the location of the source file in the C runtime. This function is at the time Microsoft built the library.
The assertion indicates an out of bounds memory access. The cause of the error appears to be incorrect allocation of topologyOfKernels.
corruption.topologyOfKernels = new kernelTopology_ [numberOfLayersInput - 1];
should be allocating numberofLayersInput elements.
corruption.topologyOfKernels = new kernelTopology_ [numberOfLayersInput];
Hello everyone and anyone!
Ok... I have been banging my head against the wall with this issue for literally weeks, but I still cannot find an answer that has successfully resolved the issue.
I created an FLA and placed a FLV component with the instance name of videoPlay on the stage.
videoPlay is pathed to a streaming FLV with embedded event cue points. The cue points are numbered sequentially from narration1 to narration16.
I established a listener object:
var videoPlayCuePointListener:Object = new Object();
The event listener for the cue points:
videoPlayCuePointListener.cuePoint = function(eventObject:Object):Void{
if(eventObject.info.name == "narration1"){_root.cc_box.cc_txt.htmlText = cueTxt1);}
else if(eventObject.info.name == "narration2"){_root.cc_box.cc_txt.htmlText = cueTxt2);}
etc, through narration16 }
and so on through narration16.
Then I attached the event listener to the FLV component on stage:
videoPlay.addEventListener("cuePoint", videoPlayCuePointListener);
All of this works very well. As the FLV plays, each event cue point fires off the correct text to the cc_txt dynamic text box.
The issue I am having is that I cannot find the nearest cue point to the FLV playhead so that I can fire events when the user scrubs the timeline.
I have researched this as thoroughly as I possibly could before finally deciding to post the issue, but although the documentation and various postings regarding findNearestCuePoint discovered throughout the web have provided numerous examples, not a single one has been successful.
I have attempted to add a listener to videoPlay that creates an object (nearestCue) and gives nearestCue the value of videoPlay.findNearestCuePoint(videoPlay.playheadTime), then read out nearestCue's name, label, etc. No dice.
Nothing suggested in any posts I have reviewed (many, many posts) has provided an answer.
This seems like it would be the easiest thing to accomplish but I have not been successful a single time.
Any suggestions or assistance would be much appreciated.
Thank you for your time!
Haven't touched AS2 in a long time. I've done a basic test and findNearestCuePoint worked. You're using the FLVPlayback component, right ?
Here's what I've tried:
videoPlayer.autoPlay = false;
onEnterFrame = function():Void{
videoPlayer.seekPercent(_xmouse/Stage.width * 100);
videoPlayer.play();
trace(videoPlayer.findNearestCuePoint(videoPlayer.playheadTime).name);
}
The recommended way would be to find the nearest cue point in an playheadUpdate handler which is triggered after the playhead changes it's value. (e.g. 1. tell the playhead to move, 2. the playhead actually changes the value, 3. the playheadUpdate gets called)
Here's a more basic approach:
onEnterFrame = function():Void{
if(videoPlayer.metadata) trace(videoPlayer.findNearestCuePoint(_xmouse/Stage.width * videoPlayer.metadata.duration).name);
}
In my test I've added 4 cue points. Tried them all: actionscript/event/navigation.
The strange thing was when I tried to access the cuePoints property through videoPlayer
or through videoPlayer.metadata I got an array of 8 undefined objects, and the length of the array was 4 when I traced it. Don't know what the issue is, maybe encoding/codec and as2 compatibility, not sure.
Anyway...as long as you've got your cuePoints array, you can manually find the closest one by looping though all of them and getting the smallest absolute difference between each cue point time and the current time:
function getClosestCuePoint(cuePoints:Array,time:Number):Object{
var numCuePoints:Number = cuePoints.length;
var minDist:Number = 100000000,result:Object;
for(var i:Number = 0 ; i < numCuePoints ; i++){
if(Math.abs(cuePoints[i].time - time) < minDist){
minDist = Math.abs(cuePoints[i].time - time);
result = cuePoints[i];
}
}
return result;
}
Here's a mockup example: let's pretend some boxes on the screen are the cue points and the _xmouse position would be the playhead time. Try this in a new document:
//fake cue points
var numCuePoints:Number = 5;
var cuePoints = [];
for(var i:Number = 0 ; i < numCuePoints ; i++) cuePoints[i] = {name:'narration ' + (i+1),time: 10 + (80 + Math.random() * 20) * i}
//visual hint - separated from the cue points
for(var i:Number = 0 ; i < numCuePoints ; i++) drawBox(this,0x009900,10,15,cuePoints[i].time,Stage.width * .5);
var playhead:TextField = drawText(this,'playhead');
//playhead update
onEnterFrame = function():Void{
playhead._x = _xmouse;
playhead.setTextFormat(new TextFormat('Verdana',11));
playhead.text = 'time: ' + _xmouse+' / cue ' + getClosestCuePoint(cuePoints,_xmouse).name;
}
//find the shortest marker within the shortest distance from the current value
function getClosestCuePoint(cuePoints:Array,time:Number):Object{
var numCuePoints:Number = cuePoints.length;
var minDist:Number = 100000000,result:Object;
for(var i:Number = 0 ; i < numCuePoints ; i++){
if(Math.abs(cuePoints[i].time - time) < minDist){
minDist = Math.abs(cuePoints[i].time - time);
result = cuePoints[i];
}
}
return result;
}
//utils
function drawBox(target:MovieClip,color:Number,width:Number,height:Number,x:Number,y:Number):Void{
target.lineStyle(3,color);
target.moveTo(x,y);
target.lineTo(x+width,y);
target.lineTo(x+width,y+height);
target.lineTo(x,y+height);
target.lineTo(x,y);
}
function drawText(target:MovieClip,name:String):TextField{
var result:TextField = target.createTextField(name,target.getNextHighestDepth(),0,Stage.width * .5-20,100,20);
result.autoSize = 'left';
result.border = true;
result.selectable = false;
return result;
}
HTH
George,
I believe I discovered the issue, and I think it was something that you covered in your previous post, but I glossed over it by accident.
The F4V I was working with had the cue points embedded using Adobe Media Encoder... and that was the entire issue.
I went back and exported the cue points out to XML, then stripped them out of the F4V and re-encoded it without them. Then I edited the XML to change all of the event cue point to actionscript and imported them into the FLA file using the FLV component properties dialogue.
Presto, amazingly enough, I had no issues finding the cue points, tracing them, and using them for any purpose.
So in the future, I just need to remember to set up the cue points in the Properties dialogue and set them to actionscript and I should be golden. So far, it has worked flawlessly with all of the F4V files since making the change.
Thank you very much for your detailed response and your follow up!