Media Asset Management
Media Asset Management (MAM system) enables the implementation of the complete workflow from media collection to conversion, storage, analysis and distribution.
Workflow of Media Asset Management
'MAM system' is not just a web server, but it controls the workflow that begins with ingesting media, processing metadata, and managing the entire process from secure storing to distribution·utilization.
Geminisoft’s ‘MAM System’ is a product that incorporates the efforts of research and development by the development team from start to finish, and is widely used by many domestic and overseas broadcasting stations and government offices.
Geminisoft provides a system that can easily ingest various types of media.
In addition, it provides an interface to easily monitor and check the content currently being recorded with a timeline-type media ingest function.
Various ingest modules
A variety of Ingest Modules are available, including modules using IO boards such as commercial video servers and DeckLinks, File Ingest Modules, and automated watch folders.
Proxy creation and media conversion
In general, original archive videos won’t be played in a web browser or will cause a heavy load on the network due to capacity problems. Geminioft's transcoder automatically generates high-compressed/low-resolution proxies for the verification of original videos.
Proxy can be created in less than 10 minutes from 1-hour video
Latest AMD EPYC Processors·
Intel Gold Xeon Processor·
7 to 12x performance with NVIDIA GPU
Proxy can be created within 5 to 9 minutes from 1 hour video
Extraction of video key frames and creation of storyboard
Extracts the image at every point where the scene changes within the video or at each specified interval. By showing this, it helps users to easily understand the entire video.
Support for metadata input for each section
Scene transition points are identified and each section is created as the starting and ending points, allowing metadata input for each section
By entering metadata for each section, it is easier for users to find the section that contains the content they are looking for, increasing utilization
It analyzes the scene of the video to locate the point where the scene changes (Scene Change Detection), and then extracts keyframes and registers the information of the detected scene to the MAM system, provide a function to easily identify the contents of the video information.
You can automatically input metadata for the section using Vision AI connections from Google and Amazon. In this case, the cost of the cloud the cost of the cloud can be charged as the amount that you use.
Search and metadata
> You can search for videos like general web searches.
> You can play the video, and check and edit the metadata if you double-click the search result.
Geminisoft was the first in Korea to use a web browser as the MAM interface in 2006.
Currently, MAM browsing with the function of MediaDirect which performs better than the web is available on editing equipment, Macintosh, and PC.
Operations of standard web browser
<MS-Windows, Apple Macintosh, Linux>
Videos can be searched for and played without any additional programs.
Support various metadata formats
Single-line text / Multi-line text
Checkbox / Combobox
Date, number, etc.
Newly added metadata is automatically linked to search engines
User-defined content can be created by type to suit the characteristics of each site, and metadata for each content type can be defined.
In addition, each metadata schema can be defined for a variety of content characteristics and immediately incorporated into the system.
Administrators can dynamically implement metadata schema
Define customer-defined contents
Define metadata for each content
Configure metadata properties
Production and Editing
Adobe Premiere MAM Custom Panel function
We provide plug-ins to connect with major NLE editors used around the world, such as ‘Adobe’, ‘Apple’, and ‘GrassValley’
The plug-in allows users to easily upload edited videos to MAM without the need for other applications.
Apple Final Cut Pro Plugin function
Final Cut Pro X
Final Cut Pro 7
Need for archiving
The media data stored in a shared device can be available for immediate use, but over time the risk of missing data and physical failure of the hard disk as well as the risk of data loss caused by viruses and ransomware may increase together with various additional costs.
In order to prevent this, the use of storage devices such as LTO for archiving can be an option to consider. But it needs a high initial cost and is difficult to build.
IMArchive (Intelligent Media Archive)
IMArchive is an LTFS (Linear Tape File System)-based archiving solution that provides the ability to archive and retrieve data to LTO and ODA storage devices to keep MAM content secure for a long time.
IMArchive main features
Restful API is provided for external integration to archive, restore, and check job status
Support for various video input formats
Checksum function for data integrity
Various tape library control
Archive management solution IMArchive performs archiving/restore operations on long-term storage devices such as LTO and ODA, and provides monitoring screens for administrators.
Integration with other 3rd party archive solutions
Spectra Logic BlackPearl
Support for high-speed file storage
Support file-based storage: NAS, Storage Area Network (SAN), and iSCSI SAN.
SAN storage and iSCSI SAN need to be formatted by a dedicated file system such as SNFS, BWFS, etc.
You can control user access to the storage that supports Active Directory (MS) and Open Directory
Support for Internet and cloud storage
Supports AWS S3 storage and various types of object storage. Object storage is more secure against ransomware and viruses than normal file storage.
Backup by workflow
MAM workflow helps you back up your stored media data, ensuring the stored media data is more secure.
The media has a bigger capacity than typical documents.
Thus, it is very important to reliably keep and back up the media data in storage.
Data stored on RAID does not mean that all data is secure and accessible quickly.
Geminisoft provides consulting services for the best storage configuration when deploying MAMs.
Integration management with Cloud·AI·STT system
Integration with Naver·Google·Amazon AI SaaS
Rather than building AI technologies including speech recognition (STT), it may be advantageous in terms of AI analysis quality to use a cloud service that continuously adds artificial intelligence analysis data.
Geminisoft provides voice recognition and voice metadata extraction functions in conjunction with AI such as Naver·Google·Amazon.
Geminisoft provides a turnkey system service that will cover deployment, contract, and payment for AI analysis.
Speech To Text, STT
By the use of Geminisoft's STT engine, STT integration service can be provided for Naver, which excels in Korean, and Google and Amazon, which excel in English and many other languages.
Depending on the results of the STT performance evaluation, the STT engine can be changed according to the administrator's settings.
Geminisoft uses Google Translation AI, providing an automatic translation of metadata or subtitle data.
It can be used when the contents would be distributed worldwide.
Video Analysis (Vision AI)
Video analysis functions are provided by Google's Vision AI and Amazon's Rekognition, automatically generating metadata.
Geminisoft can use domestic AI services such as Naver, as well as AI analysis from big foreign tech companies.
From utilizing the latest technology to solving security problems and paying costs, total services can be provided for various issues arising from using global AI and cloud services.