Time Lapse Video Options?

Has there been any discussions on Time Lapse video? I see one on Dropcam. I am looking more at using existing video survillance systems recorded video. Some options could be:

- direct connect to recorded video and output timelapse video at a user specified time interval

- exporting from a VMS video and then feeding it thru some sort of compression or time capture algorithm to product a short time lapse recording

- taking a second stream from cameras on the network to go directly into a system that can be smart at performing timelapse output.


Axis seems to offer the most options for camera based solutions, with several third party ACAP modules available. Have you seen this discussion?

Briefcam works really well, but isnt cheap. It has both a standalone and integrated into VMS option.

Sensera Systems offers cameras with cloud service that automatically produce time-lapse (in addition to video streaming).

The simplest, least expensive way may be to use a Windows time lapse application. In the discussion that A cited, one person suggested this - Video Velocity Time Lapse Capture Studio.

John years ago I saw a demo of a clever system that created a combination of time-lapse and overlay so that you could see multiple events overlaid on top of each other. Kinda looked like ghostly figures walking around the coverage area, but was cool because it allowed you to see e.g. all the people who had gone through a particular door over the last 24 hours in just a few minutes. I can't for the life of me remember who that company was. Does it ring a bell? Thanks

Something like this?

Good Choice. Vasiles, you can install the software free of charge and try it.
They have a lot of good videos explaining how things work, very customer friendly.

Should you choose to buy it's a one off software, lifetime updates are free.

Tom, there are a handful of companies who do that, the most well known is Briefcam acquired by Canon earlier this year.

Sounds like a Briefcam Video Synopsis?

Some IP cameras can FTP images up to a server if there is an internet connection. Just set the cameras upload JPEG images at an interval that suits the project, and compile them at the end of the project.

I second the Rapid Review built into NxWitness. We have used Briefcam before but the system requirements and hard costs makes Rapid Review that much better since it is built into the VMS on deployment and essentially solves the same problem.

Edit: As far as a timelampse function goes ^

Not sure you can compare Briefcam to NX rapid review. One (NX) is making a simple time lapse while Briefcam is processing the video and removing any video where objects are not moving not to mention it makes any object searchable and they can built reports on all the objects in the video.

That's like comparing a electric scooter to spaceship.

Briefcam is processing the video and removing any video where objects are not moving...

If you use NX Rapid Review on motion detected video, doesn’t that roughly achieve that same goal?

NX rapid review creates a time lapse video completely different then Briefcam.

Agreed, in general.

But as regards in particular to your statement about Briefcam “removing any video where objects are not moving”, NX RR could provide a time lapse export that would only contain video where motion is detected.

Time lapse is just a bunch of JPEG images correct? How would I see video of each motion event? Wouldn't you just see a single image or maybe 2 image from each motion event?

Just playing recorded video in the client will skip non motion evens and give you video of each event. Again way different then Briefcam.

Time lapse is just a bunch of JPEG images correct?

Let’s try a real-world example and see what would happen. I’ll just make up some scenario that hopefully is reasonable, feel free to object or give a better scenario.

Imagine we’re talking about a camera pointed to a self-service retail checkout station and we want to make an export for each day.

Lets say 16 hrs @30 fps = ~1.7M frames in total. But let’s say that only 2 hrs of the 16 hrs video has motion, which means ~216,000 possible frames to choose from. Which, say, consist of 24 checkouts averaging 5 min,

Also suppose we want the condensed video to be about 10 min in length @ 30 fps

So 10 min * 30 fps = 18,000 frames in the export.

216,000 / 18,000 = 12 frame interval

So if we make our export consist only of every 12th frame then we would get events that would average around 750 frames or 25 seconds per checkout.

Thats for a simple motion-reduced time-lapse procedure.

In what way would the 10 min Briefcam synopsis be different in the 18,000 frames it would create?

Well for starters I can search for women in a red dress with a bag walking from right to left in the FOV of the camera for the whole day. Then I could also provide a web link for the management staff on how many men, women, and children were waiting in the FOV for the whole day. I can also run path analysis can see what direction everyone walked to get to the retail checkout station.

We are talking apples and oranges here. If you want a simple time-lapse video Briefcam is not what you're looking for.

Well for starters I can search for women in a red dress with a bag walking from right to left in the FOV of the camera for the whole day. Then I could also provide a web link for the management staff on how many men, women, and children were waiting in the FOV for the whole day. I can also run path analysis can see what direction everyone walked to get to the retail checkout station.

Who said anything about search and web links, strawman?

C’mon Mike, all I objected to was your first reason for saying they were different:

Not sure you can compare Briefcam to NX rapid review. One (NX) is making a simple time lapse whileBriefcam is processing the video and removing any video where objects are not moving...

This is a waste of everyone's time UD#1. Briefcam literally removes objects from the video that are not moving and then plays all the moving objects at the same time. You can't even specify the time for the synopses as it depends on how many objects are in the video over the time you selected. You start the process by selecting cameras and a time range than going and getting coffee. Once the synopses are completed you can review the synopsis which looks like below. You can then filter objects by type, color, area, path, speed, size, direction, faces and dwell time. You can't even export a time-lapsed video!

NX can make a time lapsed of video recording based on motion detection. It's a time-lapsed video using JPEG images which you have to configure for each export. You then play these videos repeatedly until you find what you are looking for then you have to go back into the video from the camera and find time date and time the object you are looking for and then search the video.

Both are great software solutions but Briefcam is clearly better than NX when it comes to searching video.

This is a waste of everyone's time UD#1...

Please give my apologies to everyone

You're absolutely right - let me be more specific - for a simple time lapse function briefcam is like using a flamethrower when you need a lighter. As far as the rest of the analytics - Briefcam takes the cake.

(In reply to Michael Miller)