loopbio blog

Loopbio Joins NVIDIA Inception Program

Written on Donnerstag Oktober 04, 2018

We are happy to announce that loopbio gmbh has been accepted into the NVIDIA Inception Program. The program is designed to nurture dedicated and exceptional startups who are revolutionizing industries with AI and data science. The Inception Program provides direct access to NVIDIA's latest technology, deep learning expertise, and a global network of partners and customers.

Loopbio was the first company to bring easy to use deep learning based video analysis and tracking soulutions to the quantitative and behavioural biology research fields. Our revolutionary loopy product was launched in 2017 and allowed AI tracking and analysis of animal behaviour, using only your web browser and without writing any code. Loopy has been improved ever since with the addition of state of the art AI algorithms for pose and 3D tracking, and image and behavioural classification.

conda

Unlike other AI platforms, loopy does not stop at just model training. It provides comprehensive tools for performing quantitative analysis on processed video to let users get high quality scientific data faster.

About Loopbio

Loopbio was founded in 2016 to bring cutting edge technology to behavioural biology. The company is based in Vienna and provides integrated solutions for high-speed single- and multiple-camera video recording, video analysis and tracking, and virtual reality.

Motif Version 4.5 - New Features

Written on Dienstag August 28, 2018

Following the last release we have continued to add features and improvements to our Motif software. This post includes a short overview of some hightlights, while a full list of changes is provided on our website.


Each and every Motif system receives automatic updates without any extra charge.

Increased Support for Environmental Sensors

We have improved our support for environmental sensors from Phidgets greatly. This means that you can simply connect any one of their 'VINT' series sensors to a Motif system and you can now take sensor recordings automatically, at several different sample rates. Phidgets offer an enormous variety of sensors which can be used to measure various parameters of your experiment while recording video, including;

compatible sensors

A selection of compatible sensors

Because we record to our extensible and open imgstore format, all sensor recordings are immediately associated with both the time and framenumber of the video being recorded.

Controlling Outputs

In addition to measuring environmental sensors, we added the ability to switch on or off supported Phidgets outputs, relays and motors. Using our open API you can now, for example, perform tasks like the following

  • control experimental stimuli at regular or scheduled times
  • control physical devices such as motors, servos or actuators
  • switch on/off LEDs, lights or other stimuli
  • schedule tasks for before or after recording has been completed, such as automated feeding or cleaning procedures

Improved Integration with Loopy

Following on from above, if you have a recording with environmental data associated, this will be immediately visible in Loopy after the imgstore has been uploading or imported.

environmental readings showing in loopy

Graphing and export of environmental data associated with a recording is displayed in Loopy

Automatic Import

If you are running an on-site version of Loopy, your Motif and Loopy systems can be configured to allow automatic import of recordings after the completion of your experiment. This feature is especially advantageous when both systems are integrated with your IT infrastructure because all video and experimental data is automatically added to your shared and backed-up network storage without risk of deletion or loss.

loopy integration

Simply enter you Loopy username and your recording will be automatically backed up and imported


Further descriptions of the powerful integrations between Motif and Loopy will be the subject of a future blog post.

New Alignment Visualization

The last motif release added a number of image feedback augmentations to help with setting up your cameras and experimental assays. This release added a new visualization designed to help with alignment of samples inside the experimental apparatus, or to help align multiple cameras in a multiple-camera situation.

alignment feedback

The center of the image is indicated, in addition to concentric bands respecting the aspect ration of the camera sensor.

Interested?

Motif is the first video and camera recording system designed for the experiments of modern scientists. It supports single and multiple synchronized camera scenarios, remote operation, high framerate and unlimited duration recording. It is always updated and has no single-user or other usage limitations.

If you are interested in a Motif system, please contact us for a quote or to see how Motif can solve your video recording needs.

2018 Conferences

Written on Freitag März 23, 2018

Conference season is here! If you are interested in seeing our products or talking to us in person about your scientific needs you can meet us at the following places.

This post will be updated as we confirm or add more dates.

loopbio supports conferences at the beach

Motif 4.0

Written on Montag November 27, 2017

Motif video recording software version 4.0 was released today. This version features many changes and improvements, as demonstrated in the following youtube video


Motif is a video recording sytem which supports single single camera or multiple camera recording, and video recordings can be from seconds to weeks in duration. The version 4.0 release has the following highlights:

Motif 4.0 screenshots

Sample screenshots of Motif 4.0 multi-camera and single-camera web user interface

I'd like to demonstrate a the last two of these features using the new Python API.

Starting and stopping recording is easy. The following snippet demonstrates starting recording on all cameras with the predefined compression settings named 'high' (see the Web UI for the names of your configured compression formats). The recording will run for 5 seconds and will have have the additional metadata 'experiment' and 'genotype' saved to the recorded file. The snipped works equally on single or simultaenous multiple-camera systems.

from motifapi import MotifApi

IP_ADDRESS = '10.11.12.13'
API_KEY = 'abcdef123456abcdef123456abcdef12'
api = MotifApi(IP_ADDRESS, API_KEY)

api.call('recording/start',
         codec='high',
         duration=5,
         metadata={'experiment': 'frobnicate', 'genotype': 'wt'})

Likewise, the API also supports defining scheduled tasks, this allows for example, to schedule recordings and their subsequent copy to storage to occur at specific times. Task scheduling re-uses Cron syntax with some extensions (monotonic tasks).

api.call('schedule/recording/start', task_name='record_video',
         cron_expression='0 06-16 * * *',
         codec='low', duration=30*60)

We are really excited about these new API feature as they allowa laboratory wide automation of single or multiple recording systems or experimental assays. We are looking forward to extending our API support in future releases.