Is there a way to re-identify bodies, which exit the scene and re-enter it?
The SDK 2 gives new IDs on re-enter.
Is there a library for that? Or is it practicable to save body data (length of arm, etc.) and compare it to the bodies re-entering?
You would have to generate your own method of locally tracking body information on a PC the Kinect is hooked up to. You would need to set up an analysis feature where the person stands in a fixed position and either use facial recognition or measure their bodies and find the closest matching data. Then you can temporarily assign the data to the body IDs and use that to track them while they are in view.
There is no way to keep track of a body if they leave the view without going through the processes of analyzing and comparing the body or face.
Related
I am working on a project to detect anomalies in web users activity in real-time. Any ill intention or malicious activity of the user has to be detected in real-time. Input data is clickstream data of users. Click data contains user-id ( Unique user ID), click URL ( URL of web page), Click text (Text/function in the website on which user has clicked) and Information (Any information typed by user). This project is similar to an Intrusion detection system (IDS). I am using python 3.6 and I have the following queries,
Which is the best approach to carry out the data preprocessing, Considering all the attributes in the dataset are categorical values.
Encoding methods like hot encoding or label encoding could be applied but data has to be processed in real-time which makes it difficult to apply
As per the requirement of the project 3 columns(click URL, Click Text and Typed information) considered as feature columns.
I am really confused about how to approach data preprocessing. Any insight or suggestions would be appreciated
In some recent personal and professional projects when faced with the challenge of applying ML on streaming data I have had success with the python library River https://github.com/online-ml/river.
Some online algorithms can handle labelled values (like hoeffding trees) so depending on what you want to achieve you may not need to conduct preprocessing.
If you do need to conduct preprocessing, label encoding and one hot encoding could be applied in an incremental fashion. Below is some code to get you started. River also has a number of classes to help out with feature extraction and feature selection e.g: TF-IDF, bag of words or frequency aggregations.
online_label_enc = {}
for click in click_stream:
try:
label_enc = click[click__feature_label_of_interest]
except KeyError:
click[click__feature_label_of_interest] = len(online_label_enc)
label_enc = click[click__feature_label_of_interest]
I am not sure what you are asking - but if you are approaching the problem online/incrementally then extract the features you want and pass them to your online algorithm of choice - which should then be updating and learning at every data increment.
Good afternoon. I am using Agora - UE4 Bluprint version.
I can not find a way to access the frame of the new connected user. This is provided that more than two people are connected. The first connected each tick causes "OnLocalFrameReceivedDelegate"
all subsequent connected call "OnRemoteFrameReceivedDelegate". In total, it turns out that you can broadcast the image from the camera of the first connected or the second. And how to take the camera of all subsequent connected?
In our blueprint example, we have two images that receive Agora video frames:
CameraFrame - the Local user
InterlocutorFrame - the Remote user
If you call into the blueprint example with 3 or more people, you will see that you can switch between UIDs in the dropdown menu.
Switching UIDs changes the InterlocutorFrame video feed to the remote user whose UID matches the one you specify, like so:
To display multiple users, you need to place multiple UI Images on the screen, and SetBrushFromTexture with the appropriate UID.
I would suggest using the OnUserJoined callback to create a new image actor, and store the new incoming UID inside that actor. You could also store all the current UIDs in one place, or use another solution that you think is best :)
An XPage is used to display the number of points a person has collected, and the number of points remaining (see below).
I have a repeat control which gets a collection of documents meeting a specific criteria. The last column in the control contains 5 digit/form/comboboxes, which are displayed or hidden, according to the number of fields on each document that contain data.
The layout contains gift cards worth a certain amount of points, and the person can select how many of each gift card they want. eg.
Company Available in Values of Points Required Quantity Requested
The Quantity Requested column contains the digit/form/comboboxes. As the person selects values in the checkbox, I want the number of points remaining to be recalculated.
The onChange event of the digit/form/comboboxes calls a function in an Output Script which calls an RPC, which in turn calls an SSJS function. The SSJS function cycles through the documents displayed in the repeat control, gathering the points required information. I then wanted it to also grab the Quantity Requested. I understand from a previous posting that because of the way the digit/form/combox is rendered, I can only get the value using CSJS with dijit.byId and perhaps putting the value in a hidden field and retrieving it from there.
I can't seem to wrap my head around how I will do this when the repeat control will make it possible for there to be many combobox1 and combobox2, etc.
The XPage is not bound to a form, because all the items are just calculated on the fly and then discarded.
What is the best way to do this?
The JSON RPC service can't interact with any changes made in the browser, see https://www.intec.co.uk/json-rpc-service-component-tree-manipulation-openlog/. This could be the cause of your problems.
You may be able to get around it by triggering a partial refresh (POST) before calling the JSON RPC. In theory that might work, because the component tree (server-side map of the XPage) would get updated by the partialRefreshPost and the updates picked up by the JSON RPC. It's possible though that the Restore View picks up a version of the XPage other than the one for the browser, I don't know. I've never investigated that.
It's been a while since I've worked with server java script, I have been doing it the managed bean way with ActionListeners. If you have the data in the UI, then can you avoid server side processing and do it client side?
You can also use the DOM XSP Object like XSP.setSubmittedValue to have a key value pair sent with your post request to the server side, you can only have one... it can be JSON or any other value you set it to from the client side javascript.
I figured out how to do this. If anyone wants the code, let me know and I'll provide it.
We're trying to display some booking information to the users and we're asking to them the ID which has a 10 length numbers format like this one: 1553296942
In the stories, we try to identify the user input with an intent called bookingStatus and a entity called uid.
Thing is, this IDs are recognized as a wit/location type (it looks like coordinates to him, I guess) and it doesn't recognize them properly most of the times.
What would be the best approach to handle this situation?
For now, in the Understanding tab we're feeding the bot with lots of these IDs, adding the intent bookingStatus and marking it as uid entity aswell. Is this the right thing and shall we continue training it this way?
You can feed with the 10-length numbers. Actually there are only 10^10 possibilities for the uid entity. You can basically feed the whole 10^10 possibilities with a simple CURL command which contains a loop to 10^10.
How do you feed your NLP without the Understanding tab? Well..
Check the HTTP API Docs here
https://wit.ai/docs/http/20160526#post--entities-:entity-id-values-link
Have a nice day!
I'd been looking at using the Rally API to create an app to do some simple auto-ranking stories based on specific criteria (at the request of / to help out a couple of our Product Owners) in a drag&drop workspace.
I did this by adjusting the "Rank" properties of the sequence of stories (I made sure to keep the rank values within the same overall min / max range as prior to the auto-reorder). I made a call to the API to update each story's rank individually (in quick succession).
Testing the app, I found that sometimes after running the auto-ranking app, some screens (such as the kanban board) would tell me that "drag and drop re-ranking is disabled for manual rank workspaces" (and the kanban would no longer let me drag & drop, although other screens such as the backlog would still let me drag & drop, and the workspace settings were still set to drag & drop). Deleting the stories, or reordering them on the backlog screen would return things back to normal.
After trying a few ideas to solve this, I figured perhaps if I was updating the Rank on a number of stories in quick succession, the back-end might be getting confused with these (potentially several / concurrent) requests. Introducing a delay between each story's API call has seemed to avoid the problem, and to speed things up, I now update a story's rank in the API only after the previous story's rank update has invoked the "updateComplete" function.
Do the assumptions in the last paragraph above make sense based on the backend ranking? Is there any Javascript API call to update multiple stories at once? (Otherwise I'm quite happy with the solution of only calling the API "update rank" after the prior API "rank update" call has returned ok).
That sounds like you discovered a possible issue with the backend ranking. The warning you were getting stems from the board trying to figure out which type of workspace you are in (manual or dnd rank). Since that setting is not currently available via WSAPI the board tries to figure it out based on its data and sometimes gets it wrong.
Currently there is no fully supported/documented way to adjust the ranks of items via WSAPI. The Card Board component in the App SDK uses special rankAbove and rankBelow query string parameters during update calls (which you can see in Firebug or the Chrome dev tools).
How are you currently setting the Rank values?