The biggest challenge for media industries when it comes to using AI efficiently, is the necessity of training their own AI model to guarantee promising results. Out-of-the-box recognition services are often not practicable for media companies with high quality standards regarding the creation of visual content. Therefore, we have developed DeepVA, an AI platform to let you train new AI models with just three clicks and only one sample image, making AI customization incredibly easy and intuitive.
Functions
- Face recognition
- Text recognition
- Landmark recognition
- Visual concepts recognition
- Face attributes
- Diversity analysis
- QR-code detection
- Logo recognition
- Shot boundary detection
- Aspect ratio detection
- Speech-to-Text
- Speaker identification
- Face training
- Landmark training
- Logo training
- Speaker training
- Face dataset creation
- Face evaluation
- Landmark evaluation
- Speaker dataset creation