What modules are available?
Explore the Powerful Analysis Tools in DeepVA
Overview of Available Modules in DeepVA:
DeepVA offers a range of visual data mining modules that enable the extraction of valuable information from videos and images. These modules are designed for various use cases such as object recognition, face detection, speech-to-text transcription, and more.
Module Parameters
Each module has specific parameters that can be customized based on your needs. For example, you can set minimum confidence levels for object recognition or define language preferences for detected objects or text.
- Face Recognition:
- Detects and identifies public figures in categories like politics, sports, and entertainment, leveraging a pre-trained dataset of over 20,000 personalities.
- Part of the Deep Media Analyzer
- Face Attributes:
- Face Attributes recognizes emotions, ethnicity, gender or facial characteristics such as "beard", "eyes closed" or "glasses" of all persons appearing in pictures or videos
- Part of the Deep Media Analyzer
- Object & Scene Recognition:
- Automatically detects and labels objects and scenes in videos or images. With over 1,500 object classes, it helps you categorize and archive visual data efficiently.
- Part of the Deep Media Analyzer
- Lower Third Recognition:
- Reads on-screen text inserts (such as names) and associates them with corresponding faces in the video, helping in creating datasets of personalities.
- Part of the Deep Media Analyzer
- Face Dataset Creation:
- Allows for the creation of custom datasets by extracting faces from interviews or on-screen text inserts and associating them with their names.
- Part of the Deep Collector
- Speech Recognition:
- Converts spoken language into text (speech-to-text), detects the spoken language, and can even recognize custom entities from a dictionary. It also supports transcription and translation.
- Part of the Deep Media Analyzer
- Speaker Dataset Creation:
- Allows for the creation of custom audio datasets by extracting audio snipptes from interviews or on-screen text inserts and associating them with their names.
- Part of the Deep Collector
- Personal Data Anonymization
- Blurs faces and license plates in images and videos. It is a privacy protection tool that helps to comply with data protection regulations.
- Part of the Deep Media Analyzer
- Landmark Recognition:
- Identifies well-known landmarks and architectural structures in visual data, helping to classify content based on geographic and cultural contexts.
- Part of the Deep Media Analyzer
- QR Code Detection:
- Detects and decodes QR codes as well as EAN-13 codes (product barcodes) from videos and images, making it useful for media that includes scannable codes.
- Part of the Deep Media Analyzer
- Advanced Diversity Analysis:
- Analyzes the percentage of gender and age representation in visual content, which can be useful for monitoring the diversity of characters in media productions.
- Part of the Deep Media Analyzer
- Subtitle Detection:
- Detects burned-in subtitles in videos, identifies their position, language, and text content for further analysis or extraction.
- Part of the Deep Media Analyzer
- Text Recognition:
- Find and extract characters or words in a media file
- Part of the Deep Media Analyzer
- Diversity Analysis (Legacy):
- Diversity Analysis offers the possibility to determine the percentage of gender occurrence in images or videos. This module was be replaced by the Advanced Diversity Analysis and will soon be deactivated.
- Part of the Deep Media Analyzer