Slashdot Mirror
Ask Slashdot: Open-Source Forensic Surveillance Analysis Software?
McBooCZech writes "I am trying to set up a surveillance system. It is not intended to build a real-time on-line surveillance system to watch a wall of monitors on a 24/7 basis. The main scope is to record video (24/7) from the fixed cameras around our facility and when needed, get back to pre-recorded video and check it for particular event(s). Of course, it is possible to use a human to fast forward through video using a DVR-type FF function for short video sequences. Unfortunately, for long sequences (one week), it is not acceptable solution. I was searching online the whole weekend for the open source software for analysis of pre-recorded video in order to retrieve events and data from recorded video but had no luck. So I ask you, Slashdotters: Can you provide some suggestions for forensic software to analyze/find specific events in pre-recorded video? Some examples of events: 'human entering restricted zone,' 'movement in the restricted zone,' 'light in the restricted zone.'"
Something about an autoturret system for shooting squirrels with a watergun comes to mind...
Zoneminder.
http://www.zoneminder.com/
ZoneMinder is highly componentised and comprises both the back-end daemons which do the actual image capture and analysis and a user friendly web GUI enabling you to both monitor the current situation and view and organise historical events that have taken place.
http://www.ispyconnect.com/ Open Source Camera Security Software
http://code.google.com/p/openvss/ Open Platform Video Surveillance System
http://www.zoneminder.com/
It might be a lot easier to install sensors to log interesting events that you can go back and review the video.
Some examples of events: 'human entering restricted zone,' 'movement in the restricted zone,' 'light in the restricted zone.'"
Just tell Homeland security that some occupiers are planning a protest. They'll pay for the install and maintenance of your system. Every now and then, leave a deflated half-assembled tent in the parking lot...
#fuckbeta #iamslashdot #dicemustdie
Many commercial DVR/NVR's will have this functionality built in. Motion recording, motion event triggering, motion search, and motion search by area are rather common features in commercial CCTV software. I have never seen anything approaching this in open source or free software.
There are several brands I would recommend. Any of these can sell standalone servers or just the server software.
Exacq (www.exacq.com)
milestone (milestonesys.com)
Avigilon (avigilon.com)
Of those three, Avigilon has the better video handling, IMHO. Especially when working with 3, 5, or 16MP cameras.
All are somewhat comparable in price.
http://www.lavrsen.dk/foswiki/bin/view/Motion/WebHome
Google you retard!
I'm sure somebody here will help you with this. They probably shouldn't.
Help stamp out iliturcy.
I'm assuming what you mean is you want to be able to detect motion in a real time feed. You may be able to cook up some voodo with "Motion". http://www.lavrsen.dk/foswiki/bin/view/Motion/WebHome Although it may take some modifying because it is meant mainly for usb webcams, etc.
Hire 3 independent sets of "watchers" so you can have more confidence that events of interest will be caught. After a while you might be able to let go one or two, if they always have the same hits...
This issue is a bit more complicated than you think.
If you have a Mac and some cams (up to 60 cameras)
http://www.bensoftware.com/securityspy/
and http://www.bensoftware.com/securityspy/features.html
"SecuritySpy can send email notifications, play alarms, or run scripts when motion is detected."
Domestic spying is now "Benign Information Gathering"
Video analytics can only help out so much.
Realistically, video analytics can and often does present a lot of false positives, so its not the be all and end all to security.
I would not rely solely on CCTV for any half decent security solution, purely because it only offers after the fact protection.
What I would be asking is what events are you specifically looking for ?
I would honestly suggest you use CCTV as part of a security plan, and not using it as your only plan.
Access Control systems can log who, when and where people have attempted to and gained access.
Alarm Systems can detect motion in areas and respond based on arming status and schedules.
You kind of get the idea.
Using other methods of physical security can help you determine when events may or have taken place.
This in tern aids you in searching for video footage, as if you know roughly when an event has taken place, searching through a couple of hours of video footage for one or a couple of cameras, (At 8x, 16x or 32x speed) becomes a whole lot easier than trying to get the perfect video analytics configuration that wont waste your time with thousands of false positives that you will need to watch for a weeks worth of video footage.
If your after simple CCTV security, and you dont have a lot of traffic (pedestrians, vehicles or other movement), then motion detection can also make life
a lot easier, as it tends to cut out a LOT of video footage during quiet times (Night time for example).
Multiple webcams:
http://ubuntuforums.org/showthread.php?t=1897786
Motion can be told to act on or ignore a light switching off/on using a threshold.
It can be setup to detect large or small amounts of motion and long or short periods of time before triggering.
I'm not sure about singling out specific parts of the screen as triggerable/not triggerable.
The government which is strong enough to protect you from everything is strong enough to take everything from you.
I agree, Zoneminder is a mess. Although, potentially powerful enough for a large system. For smaller stuff, I've used wxCam (which supports motion detection, supposedly) but the image (or my webcam) seemed to darken/degrade over time.
http://wxcam.sourceforge.net/
Clearly out of price range, however this is a pretty cool solution: http://www.boeingvsoc.com/. Does everything from 2d and 3d modeling to multi camera angles on a building.
Hire a fourth watcher to watch the watchers, in case they sleep or masturbate.
Pay this watcher $1 more per day.
What I want to know is where I can get the one that allows you to zoom in infinitely and refine a sharp image out of a few pixels.
What you are asking for is a computer vision system. Typically there are parts, but no complete Open Source implementation of what you want, unless you are willing to assemble them yourself, and accept somewhat less than commercial quality.
Here's the most comprehensive resource: http://www.roborealm.com/links/vision_software.php
Movid is the part you want for human tracking; typically these systems are going to require parallax cameras, meaning binocular vision, for some of the recognition.
Try training a neural network to recognize patterns.
You could also have a rotation of shifts of humans responsible for each segment and have them reviewed in parallel.
VLC does motion detection. Tools > Video Effects > Advanced > Motion detect.
I haven't really figured out how to use it, like stop when it detects something. One thing that could be better is play speed. It goes to 16 which is kind of slow for a weeks worth of video.
WTF is wrong with the editors, this is an obvious troll post with bullshit question. In about 3-seconds it takes to type open source video recording you get all the answers out of Google. Douchebag poster and double-douchebag for the editor who approved this non-story.
Slashdot is now a corporate non-story advertising medium full of shills and trolls posting fake questions to associate with text source based ad-revenue spam ads.
Zoneminder + Zones + Timeline
There's a cool trick called "background subtraction" that lets you watch only the pixels that are changing in a video. For an example, check out this open source software: Scene.
Mobotix camera (which run linux internally and have control software that runs on linux) will record video as variable rates dependent upon whether there is motion.
So you can have 1 frame per second when nothing is moving and 30FPS when there is motion within the frame (or a specified zone within a frame).
When it comes to playback, due to the reduced frames it effectively fast forwards for you. You can also choose to only record when there is motion (with a definable buffer for before and after).
If you still have to buy the cameras anyway, look for camera that support the features you need out of the box.
You can probably do the same thing with cheap basic camera and open source (free as in beer) software. The value you place on your labour will determine which is cheaper.
Put a bunch of guys in scramble-suits and have them fast forward and rewind the tapes until they go apeshit and start ingesting massive amounts of mind altering chemicals. Then... profit.
There are a million and one off the shelf solutions that will detect movement, etc and trigger recording based on it and likely something came with the cameras. I don't know of any that both record 24/7 AND highlight events. It might be possible to duplicate the video stream through two solutions or have overlapping coverage cameras with some streams set to record 24/7 and some that trigger on movement.
The easiest way is to combine it with running an event based system in parallel such as the hydra control freak. This will let you record just the events based on
sensors placed in the correct places external to the camera. You can then view just the events very easily and if one warrants further investigation relate the time back to your 24/7 based recorder. Using PIR, IR beam or other systems in the protected zone will provide much more reliable intruder detection that trying to do video content analysis, particularly at night.
...recording a *RESTRICTED* zone!!!
(From summary: "Can you provide some suggestions for forensic software to analyze/find specific events in pre-recorded video? Some examples of events: 'human entering restricted zone,' 'movement in the restricted zone,' 'light in the restricted zone.'"")
Try the software from RetailNext. They have a very robust analytics software, which can do Dwell Analytics, People counting, Face Detection , Gender Recognition etc..
Kestrel. Nuff sed.
Excuse me, but please get off my Pennisetum Clandestinum, eh!
Buy a Canon camera and use a CHDK motion detection script to only record video while there is motion in the field of view.
I think this is the base level functionality for most systems. They'll only record information when the lighting in the frame changes or something passes in front of the camera. Even your basic Swann security system will do that, and let you watch highlights from your iJunk. Typical on Enterprise grade CCTV products as well, with export options to various USB devices or email. Better packages will print activity reports as well. Dalmiere is another Enterprise grade device I've used - plug it in with its cameras and let it do its thing. Records about 2 years worth of data from about 8 cameras at a time. Remote access to the box, etc. We'd shoot through days of footage in the space of a few minutes, and most places will have security systems in place to give you a better idea if your perimeter has been compromised.
My QSee from Costco has zones that you set up by drawing a mask on a grid atop the video for each camera to show which spots in the view that you care about, such as the sidewalk on a grassy lawn or front door on a house, and when something comes into that field of vision, the DVR marks a list of "events" where something entered that specific zone. You search by events per camera, not by human recognition or anything fancy, so motion triggers an event and in quiet areas there are only a handful of motion events per day. It also allows an alarm or other device to trigger the event, such as a door opening I guess.
Sometimes when the trees sway, the sensor sees an event when nothing happened, or if a spider builds a web in front of the camera or a moth flies by, it shows a false event.
QSee's software sucks overall (hard to remote access using an IE8 ActiveX control for example- DVR uses a closed Linux distro I think), but it just works most of the time for local connections and doesn't crash.
At home I have a camera in the backyard. It detects motion but due to all the waving plants and cats strolling around I get thousands of pictures per day.
So I wrote a quick hack using opencv which detects faces: http://www.vanheusden.com/detect_faces/
That brings me back from thousands to tens of images to verify.
http://www.aimetis.com
It's great entry level software with decent analytics. It's not going to pull CSI Miami stuff but it will do what you want
By "Forensic", do you mean accepted by the courts? If not, what does "forensic" mean and what does it have to do with a tool's technical capabilities?
And do somone have forensic tools for web pages? Wit which i can see who acces to my pages?
Web games http://robotower.com/
They have very nice camera's and the software you get for free with it.
http://www.mobotix.com/content/view/full/1723
take a look at vitamindinc they have a free version of their recording viewing software which I believe allows you to scan video for events. www.vitamindinc.com
Why are you going back in time for analysis?
There's a plethora of products that record (DVR-style) with analytics for the stuff you're asking about, but they all do it in real-time as it happens. Milestone comes to mind, as do Panasonic's offerings.
Why do they work this way? Because if it were that important, you'd want to know right away, not a week later when you get around to it.
Failing that, if visually scanning a weeks' worth of video is taking too much time to be worthwhile, then perhaps you need to re-evaluate what the incident is worth.
Kid-proof tablet..
I see, so when you say Open Source what you really mean is Free because this has nothing to do with using and contributing something back. Otherwise you would have said so.
For example, you could donate to the programmers who make the software that you decide to use. You could have an interesting custom requirement that you could make a module for and release as OSS. Or you could find and fix bugs or documentation for some existing package. Why don't you offer to do something like that in your question ?
I use vitamin D software to monitor the webcams in my apartment. http://www.vitamindinc.com/ It does a pretty good job detecting whether something is a person or not, and you can configure it to send you an email when it detects something. It also only records video surrounding an event. The starter version is free and has some restrictions in terms of total cameras, but its not that expensive overall. Presumably a place that has a "restricted zone" has a security budget more than $0.
It is not open source but I have been using Blue Iris http://blueirissoftware.com/index.html.
It does motion detection, hot spots, snapshots, archiving and email alerts.
You could set it up to record video all the time (if you want) and have it save a snapshot when someone enters a hot spot.
Then you could look through the snapshots to find suspicious activity and use the time displayed in the snapshot to locate the event in the recorded video.
The price is pretty reasonable.
Try BriefCam, not open source but it looks like what you are looking for. www.briefcam.com
Hire a lot of unemployed and offer a bonus for finding something.
- I've got bad karma because I won't parrot everyone else's opinion
huperlab.com offers excellent analytic tools with their systems. Most of the issues you mentioned are covered by their systems. I know you're looking for a cost effective way to manage your surveillance, but you gotta pay to play.
There is a product called Vindex which is specifically designed for "indexing" video. It intelligently collects up "events" and displays a gallery view (bunch of thumbnails) representing these events. By "intelligently" I mean that it isn't just looking at raw differences between frames but does considerably more math to eliminate false positives.
It can also process video at up to 60x normal playing speed, which means analysis of a night's worth of video takes a few minutes. Compared to just playing at high speed you don't miss very short events which are way too easy to miss with high speed playback.
No, it isn't open source. There is a free evaluation available at www.infinadyne.com so it is easy to check out. There is a new version in the works which will add video enhancement.
That would be a pretty good idea. I'd use it and stop pirating stuff (kidding).
Okay, so yet again, another Slashdot question along the lines of this:
"I have something that I want to do that is demanding, and which at its core uses technology that had to be specially developed to fill a niche. But I don't want to pay for it." *waves open-source flag* "How about something open-source? Can you guys do the legwork for me and tell me how to get this for fre*backspaces a lot* using open-source?"
I favor open software and open standards. But don't you think that if it takes a bit of hard work to come up with a solution, the people who did the work should get compensated? These questions...the kind I'm referring to...are not related to commodity items like firewalls or an operating system or a text editor. They're all niche specialty functions with unique and significant challenges. Slashdot should stop encouraging the monkeys who care more about getting something for nothing than they do about open source.
For your security, this post has been encrypted with ROT-13, twice.
I'm not entirely sure that what I'm about to recommend will hit a home run for you, but I wanted to throw it on the table anyway so you can review your options and see what works for you.
I run video surveillance at home. I have two outdoor cameras mounted in hidden locations which watch the front door and back door of my house. (my house is on a corner, so in some way shape or form both the front and back door is visible in some way from the street) I did this after a package went missing on my door step that I was expecting. Since I often do I.T. side work it's not rare to have rather expensive parts on the door step, and due to my work hours it's beyond a hassle to require signatures. I run a Linux server at home (Ubuntu Server 12.04, but most major Linux distros should work fine). I run a package known as Motion, which is in the repos. Motion has no GUI, it's a daemon based solution which will watch over the MJPG stream of each camera and record snapshots accordingly to what the preferences say. I won't speak too much about setting up Motion because I actually did a YouTube tutorial on it, so if you like, check it out and see if it fits the bill. http://www.youtube.com/watch?v=NwDLkMPLTw0 (there's also a part two)
Motion works with MJPG. That means it does not take H264 streams. MJPG is not going to give you a high definition amazing super awesome stream @ 300 FPS that's absolutely flawless. MJPG is a series of JPG snapshots. Some of these newer systems are all about H264 because it's more bandwidth efficient when you're doing higher FPS. If you're doing this for video surveillance, the reality is all you need is one good snapshot to catch the face of a crook, so running MJPG at 2 or 5 or 7 or 10 or whatever FPS might be a very likely solution. If you were running a network of traffic cameras or perhaps Vegas casino cameras, having a higher FPS to watch the entire stream of each step might be more warranted. Motion admittedly requires a little bit more setup because Motion isn't based on a database where you can select "play video from yesterday" and it magically finds it and you can chow down on a donut while viewing the feed back. Instead, you have to set that structure up. You have to create directories for each camera, you have to set the mount points of the drive to be used, you have to organize it as you see fit. I can tell you this, if you get through the setup stage and begin to use Motion, you'll find that Motion is absolutely bombproof. I've had many extended power outages where my server went down. Each and every time, I fire it back up and it just worked perfectly. It may not be packed of a ton of pretty polished GUI features, but I can tell you it's reliable as hell. After about a year of using it, I really can't remember a single time where it gave me an issue.
You can also set up your own web page like I did, allowing you to see a montage of your available cameras and then click on each one to get a full screen view of it. I posted some updates to this FAQ to help users with that... I personally use Method 2 - http://www.lavrsen.dk/foswiki/bin/view/Motion/FrequentlyAskedQuestions#How_do_I_see_more_than_one_camera_stream_at_a_time_63
I just keep that local "motion.html" file saved and bookmarked. That way I click it and it just brings up file:///home/jason/Documents/motion.html or whatever it is as if it's a real web site. It won't be the prettiest thing in the world, in fact it'll look extremely bare, but you'll be able to get a glimpse of what's going on. Keep in mind FPS of the webserver playback is independently controlled from the FPS on the cameras. You can record at 7 FPS or whatever and have the live view of the montage be 1 FPS, etc. A substantial amount of businesses aren't equipped with tech savvy people, so having a DVR you drop in place and plop the cameras on their mount points makes more sense in some cases. For me, I'm all about open source software and I'm all about software that does the best job that I can manipulate in many different ways. Motion gives me t
Hello,
In short if you don't want to keep reading: OpenCV + IP cameras (and a lot of work).
Now if you actually want to keep reading:
I am Leonardo M. Rocha and I have worked as software engineer in the research area of in this kind of systems for 3 years in this team: https://team.inria.fr/stars
You can check some publications on the subject here: https://team.inria.fr/stars/publications/
There is no right answer, it all depends on your particular system, what it actually has to do, the light conditions, the areas, the events, and the type of things you are actually observing. The best and "easiest" solution would be the following: hire professionals ( check here http://www.itea2.org/project/index/view/?project=1142 for some of the big players in europe), and be ready to pay. If that is not an option, you can start playing with IP cameras (or CCTV, all depends on the legislation in your country if you actually want the videos to be able to be used in legal issues). IP cameras come with some nice features like motion detection.
Then check OpenCV http://opencv.org/ and the wiki http://opencv.willowgarage.com/wiki/ be sure to get into the yahoo mailing list too. And start using whatever you find usefull there. Do not hope for too much, it is not easy to use and the documentation is actually bad (but at least you can find some useful things out there), but is the BEST library available for the public out there.
Some of the "simple" things you could do is to record only when there is movement, search for the faces, extract them and send an alarm and show the results to an observer who should make the actual evaluation. I have built some small demonstrations with this concept for this paper: sadio.opentierra.com/SADIO-Files/Girgit.pdf
More complex things, like detecting luggage (one of the things we did for this http://www.itea2.org/project/index/view/?project=1142 project), tracking, gestures, actions and other things are really a pain, need a lot of tweaking and have to be adapted to the actual scenarios. Light, contrast, definition, movement of courtains, trees outside the windows. Real time processing IS an issue, you need good algorithms and good architecture. All complex models NEED a 3D calibration (to be able to map 2D to 3D) and they WILL make mistakes (ther is incomplete information for the mapping) so you MUST make assumptions for the calibration.
If you can work with the kinect or some similar 3D cameras, that would be awsome, they take out many of the 2D camera issues (depth makes tracking and segmentation easier), use either microsoft's SDK or you can try http://www.openni.org/ an open source version that we have managed to make work under Linux and OSX.
Binocular cameras ( I haven't used ) are a pain, calibration for the 3D is not easy and are most used in mobile and autonomous robots. Maybe you can check for the subject, but I do not recommend this as starting point.
Things you WILL have to know to be able to do some thing:2D and 3D geometry (calibration), be able to use artificial intelligence algorithms (not an expert on the algorithms, just understand enough to use them), heavy C++ coding, Python, compiling, configuring your computer, manage your network (if you are going to use HD cameras, and many of them, that WILL saturate your bandwith at some point, specially with wireless cameras). And overall, you will need time and patience.
Best of luck
Leo
I use ZoneMinder with most cameras set to 'Modect' so video is saved only when motion meeting my criteria has been detected.
ZoneMinder presents an array displaying the number of events recorded by each camera in the last hour, day, week and month.
If you also kept your 24/7 recordings you could use the ZoneMinder events to pinpoint where/when to go look in your video archive.
No brain, no pain.
If you own a computer with a webcam, or (seems more likely) own usb cameras or network cameras, the software "motion" does the trick for me. It can capture frames only when motion is detected in the "restricted zones" or all the time or at given time intervals. It is available in the debian software repository and can be installed as easy as apt-get install motion. Im assuming its just as easy for other distros.
Here is a product that is used for keeping people away from industrial machines:
http://newtonlabs.com/machviz_quadcam.html
I've some professional experience with camera systems and I can echo/add the following - 1) use multiple types of sensors. Zigbee based devices, or for a larger office building type scenario something like ubiquiti's mFi. Most cameras have I/O and many have an SDK you can mash up something that fits your exact needs but watch out for 2) all these camera companies have complicated learning curves for their hardware, few on-line support communities compared to other hardware categories I work with, little documentation, few reviews, and pretty insane price levels for their digital offerings which brings me to 3) analog cameras have enormous value, if you do some careful shopping you can find good resolution and features for 1/4th the price of digital - I like CnB's products but there are others equally good. If you go analog you won't get on-camera analytics and will need some additional hardware on the DVR side.
for digital cameras look at Vivotek for value, and Axis for very professional 'just works' functionality. Axis' software and firmware feels properly engineered and well designed, something I can't say for any other brand I've worked with. They're 30%+ more expensive than Vivotek and similar /digress.
on the dvr side, you have to match the software to your needs, there are no 'great' packages out there. I've had good experiences with 3VR, who also have a decent SDK, but they're Axis-level in price. Identify a few of your specific needs and keep a tight focus on just those and you'll find a decent solution. Think hybrid with other types of sensors and keep a level, practical head and expect a lot of effort and you will do well, otherwise find a security systems integrator :)
closed minded is as closed minded does
It's not quite clear to me what you want to do... but here at work, we have a number of cameras in our server rooms, um, sorry, "computer labs", running 24x7x265.25, on servers running Linux. On that, we have motion taking the feed, and when motion is detected, whether it's people, or the lighting blinking, or too many clouds alternating with sun, and it sends the video in an email.
Will that do what you want?
mark
Using motion: http://www.lavrsen.dk/foswiki/bin/view/Motion/WebHome you can define restricted areas, define amount of change in the field of view to trigger some kind of alarm, define special commands that are triggered when uploading video stream, etc. Quite versatile I'd say.
#
#\ @ ? Colonize Mars
#
Is installation of cameras just plug and play? I've heard that you have to mess around with baluns. Is that right?
Also, what cameras to get? IR? Dome cameras or bullet? What mm should they be? How do you know?
What about viewing over the Internet? Also, how easy is it to hack them? Any additional security recommended?
How far should the low-voltage camera lines be from medium voltage (120 and 220 volt) lines to not cause interference?
I'm not a lawyer, but I play one on the Internet. Blog
I posted this anonymously but now I can't find it on the listing here. If this is a repeat from another post (that I cannot find), I apologize. I'm not sure if my thoughts here will be directly in line with what you want, but I wanted to share what I did in case it would help.
/media/surveillance/front_cam/snapshots. On top of that, I also run 24/7/365 full time recording, which utilizes the H264 stream of my cameras. Motion does not support H264, so this particular stream is not involving Motion at all. Motion is only dealing with MJPG. My cameras have
To start, ZoneMinder is a project that I have a soft spot for. It's a very complex application that works moderately well than even some brand new proprietary solutions out there today. That said, ZoneMinder has to go through a monumental facelift before I would consider using it again. Last I used it, it would run for a few minutes then just stop recording and populate thousands of errors in my syslog. ZoneMinder has had some recent developer activity with some individuals who are taking it upon themselves to do work on it and patch it accordingly. The problem is, these individuals don't have access to patch the core version in the repos. I wouldn't be surprised if a large update comes out soon or that ZoneMinder gets forked under a different name with all of the updates. Either option is possible. That said, ZM isn't my software of choice at the moment.
Instead, I use Motion. Just a quick disclaimer, Motion is admittedly not for Grandma Edna who never used Linux before. It's something that requires a little bit of setup. It's a daemon, not a GUI. You have to create the directories for each camera feed. You have to adjust all of the parameters you want within the config file. You have to lay down the initial foundation upon which Motion will run on. If you want some insight with getting started, have a look at my YouTube tutorial I set up a few months ago. http://www.youtube.com/watch?v=NwDLkMPLTw0 (take note there's also a Part 2)
Motion works by watching the MJPG stream of your camera and taking snapshots accordingly. Those snapshots can be configured to stay regular JPG's, or you can turn off JPG's all together and have the system stitch them into an avi file and have more of a video-esque playback. Because MJPG isn't that fantastic with compression, it's literally impossible to have a flawless amazingly epic jaw dropping 3000 FPS feed that looks like BluRay quality. MJPG is solid, but it's not something you would want to be running as a traffic cam where every second makes a tremendous difference. The reality is when it comes to surveillance footage, it only takes a single JPG to really capture the face of a crook. MJPG is very well suited for things like that, so running an MJPG setup at 2, 5, 7, 10 FPS, whatever it may be could very well be a home run.
As I said, Motion has no GUI. You may be wondering, but wait, I want to have a montage of all video feeds actively running. You can do that relatively easy, but it'll take a little bit of leg work. Here on the Motion FAQ I posted some ideas as to what you can use to create your own montage. This will be very basic with no additional features, but you'll see all cameras running at once. Likewise, you can click on each feed to see it full screen. I built my own motion.html file and just keep it saved locally and open it in a web browser when I want to have view. I use Method 2, but each one should serve its purpose: http://www.lavrsen.dk/foswiki/bin/view/Motion/FrequentlyAskedQuestions#How_do_I_see_more_than_one_camera_stream_at_a_time_63
I personally run dual streams. I have Motion running at 1 FPS with the MJPG stream of my cameras which saves the JPG's accordingly to the specific directory for that camera... meaning my camera out front will save the JPGs in
Try adding MTI to your search terms.