Proof Point
Recent AI breakthroughs have led to significant improvements in areas like vision, language, and gesture recognition
Selected AI Applications
Proof Point Findings
- Computer Vision – Processes and draws insights from visual content, such as images and videos, commonly used in environment sensing, navigation systems and machine-human interactions
- Sample Training Data Sets: Google’s 50 million doodles, YouTube-8M, and Megaface’s one million faces
- Natural Language Processing – Understands and analyzes natural human language through voice or text, commonly used in sentiment analysis, spam filtering and conversational user interfaces
- Sample Training Data Sets: 35 million Amazon reviews, five billion archived web pages and 1.3 million emails
- Speech Recognition – Recognizes, understands, and translates spoken words into machine-readable format, useful in machine translations, transcription software, criminal identification and voice biometrics
- Sample Training Data Sets: 630 English dialect speakers recordings and Switchboard’s 2,400 conversation
- Gesture Recognition – Interprets human gestures using mathematical algorithms, used in emotion recognition, virtual reality, and alternative computer interfaces
- Sample Training Data Sets: Affectiva’s two billion facial frames and Thumos’ 45 million action frames
