Flex :: Hand Gesture Recognition With Camera Video
Jun 30, 2011
I want to implement an flex application using actionscript 3: with the camera and the virture buttons(like a image) in the camera, one can move his hand to pass over the virture button field in order to fire the button. Now my solution is following this way: capture the button field in every frame to get a bitmapdata, and calculate the gray-scale summation of the bitmapdata, if the difference between the two frames is greater than a given value, then fire the button. Actually, it could work, but it is very sensible to the lighting condition of the background, worse, when there are more than one buttons, people often fire the button by mistake.
Has anyone worked on gesture recognition? I want to know how to proceed on comparing two similar gestures. E.g. : If A draws a circle on the screen, B should draw the same and the program should compare both the circles.
Usually Adobe Flash applications can access any usb-webcam.I have a USB camera for microscopes which understands TWAIN and DirectShow.Is there a possibility to access the video data from this camera with flash?
I have a video display in my flex app. which is attached to the camera. The camera resolution can vary depending on the user's selection, While the video display (preview) dimensions are fixed. I am then doing this: bmd.draw(vidDisplay);
Where bmd is the bitmap.. The problem is the resulting bitmap has the correct size of the camera settings (which is set) but the capture from the video is always the size of the video display inside the bitmap.. I hope i am making sense. The result looks like a little picture inside a bigger image top left justified. I want to have a video display preview with fixed height, but be able to take snapshot with varying resolution.
I am Trying to Apply some information Like text and An Image as an overlay to create a overlay effect as lice a security camera with time and date on the video along with a png based Logo.
I can record VIDEO Using Flex and FMS or any other Media Server. But I want to save a modified version of the stream being uploaded.
I am capturing camera on flex for android using air 3.1,and found that the camera is only showing the a part of of actual camera video.When i view camera using normal camera application in android galaxy tab, it shows full camera area but when i am capturing the camera in flex its showing only the center area for it.I tried using setMode to set the resolution, but its not of any use.Is there any way to get full camera area captured inside flex using camera.getCamera ?
I had an flv recording from a webcam.It mostly recorded fine, but there's one point where I paused the broadcast for an intermission.A few minutes later the broadcast started again but the video doesn't play beyond that point, only audio plays.A few minutes later in the recording the broadcast switched to another camera and the video starts playing again just fine from there.I can't seek past the frozen point either to skip a couple of seconds ahead.The freeze is at about 9:00.If I seek to 9:01 it jumps ahead 8 minutes to where the camera switches.What's strange in this is that the video is *definitely* in the file.Using Moyea's FLV Editor Pro I converted the FLV to an MPEG and the whole video is there, so it's not that the camera stopped transmitting, it's just that the playback isn't working.
I am looking to broadcast live video and need a camera that is better than a simple webcam. It looks like they used to make video cameras that would register as webcams, but stopped making them.If you wanted to broadcast an event live, what would be a good camera to use?
I have this simple code to access the webcam:[code]The quality of the video is extremely bad, with big compression fragments.I am using a built-in webcam as test, so I don't expect too much. But what I see is not normal for my webcam.So how do I set the quality of the video?I cannot see any difference in quality if I remove the setQuality call.
I would like to move (right or left according the current view) between 2 views when I move my finger on the mobile screen; for example with Google+ mobile application, in profil section, you can change the view just on move your finger on the screen, and the offset and the moving speed must be checked before change complete view.
I am trying to run a very fast camera feed. I have the fps set to 60 on the camera and I am adjusting the movie framerate to that after camera becomes active.
However, when I print out individual frames, the bitmaps seem to be the same for a bout 2-3 iterations,meaning that in the enterframe, the camera doesnt seem to have updated the video output.
Anyone on how I could make sure it refreshes the video display quicker?
I just came across the new stageVideo, as I am not very Savvy using AIR i was wondering if i could use the camera video (getCamera)on the stagevideo, and how to implement it?
I have three videos recorded from different angles of the same subject. As you can see in the attachment, I need to see one video screen with 3 buttons to jump between the different angles. The transition between the videos should be immediate.I thought of loading the videos internally, but they are over the 16000 frames so they will not load properly into Flash - so using .flvs may have to be the only option. I figured if I could get all three videos started at the same time then I could toggle the viability, but how can I get the three videos to start at the same time?
I learned from some previous post that the default encoding format used when you publish the camera from a Flash player is Soreson. I'm wondering whether it is possible to select to use other codec such as VP2 or H.264?
I'm completely new to FMS but a seasoned Flex developer. We're building a Flash Builder 4.5 application which will be deployed to iPads. The basic functional requirements are as follows:
1. Presenter in a room has a camera streaming video of themselves2. Participants in the same room have an iPad - there are 30 participants and the stream is sent to all 30 iPads simultaneously.3. There must be lowest delay possible with good quality video - we understand this can be achieved with RTMP and that VP6 has to be used (as there is no H.264 support at present in AIR on iOS)4. The presenter will trigger certain things from their iPad which will then display certain alerts, content on the partcipants iPad (from server connection the participant iPads are listening to)
Firstly, I'm interested to know your thoughts on this current spec. This was put together from zero initial experience with FMS and having read a lot online and creating a proof-of-concept. Secondly, with regards to the two-way interaction: Is this something FMS handles? Previously I have used LCDS or BlazeDS for the data events & streaming aspects of our applications. Initially I was planning on using BlazeDS again, in addition to FMS. But from what I have read FMS possible handles data push too? Does FMS have LCDS/Blaze built in in some way?
I am trying to make an application which will do the video conferencing. I have read some article on actionscript and started making the application. I am capturing image from web cam and copy it in an array which acts as a buffer. Then I am running a timer which pops each image from array and converting it into bytearray and send it via socket. When timer event is fired camera gets blocked and the whole process gets slow down. Is there anyway to do the video conferencing smoothly ?
From a Flash object, I would like to be able to detect the presence of a webcam that the user might grant access to, without actually asking the user whether they want to grant access. This is to streamline a large application that has extra features if a webcam exists — if there is a webcam, the application will present several dialogs that lead up to the normal cam-security prompt, but if the user doesn't have the hardware, I want to skip over them altogether. Requesting access to the cam before those dialogs won't make sense to the user.
It appears not to be possible to call Camera.getCamera() without the security prompt happening. Is there a workaround? (For instance, could I, perhaps, call getCamera in a hidden Flash object and inspect its return value, or does it block until the user acknowledges the security prompt?)
I have a video capture device on a machine. I have written a windows application to capture the video from this device. While capturing, if i try to broadcast using flash it does not connect. If i'm already broadcasting and then try to capture then my windows application does not get the capture pin of the device. So, flash seems to be taking the capture pin. Is there anyway to broadcast using the preview pin on the device so that i can capture using my application. Or is there way where i can split the capture pin and give one to my app and one to the broadcaster
currently I use the following workflow for Live video streaming.I have FMLE on a computer streaming the Video feed to my FMS.I currenlty use Dreamweaver to create window with a few instances of a video window and one of the windows is configured to rtmp to my server and the stream name of the stream I have setup in FMLE.This then streams to one of the windows.I then assign a different Stream name to the other windows.
I send an FMLE client to a user in the locations that want to have a video presence on the webpage.I send them instructions and running FMLE and the parameters to use to get their camera up and running.This works fine to a point.Most of the users are not technically minded and so often this leads to issues and delays to get these things to work.
I want to move on and improve this flow.I want to create the video instances on the page where:Below each video window is a text input field where a user can type in their name which will be used as the stream name.Below each video window is a button that the user clicks to connect their camera automatically without having to install FMLE.Once the video feed is running, only that user can click the button again to disconnect it. ( The button changes to "Stop Broadcast" etc).
I have looked at the structure of Ustream for example, and this seems to be the method they use as you can download the XML file of your channe,l and the RTMP URL and stream name can be copied and then used directly in FMLE and you can stream with more control of your camera that way.
