VISION AI DATA
The highest quality Vision AI training data, specializing in GDPR compliant opt-in face and voice data collection from the most challenging in-the-wild and studio environments.
Video, depth video, audio, images, multi-modal, task-based or naturalistic, cross-device or device specific, web-based or custom apps
Facial features, emotions, gender, age, actions, objects, contexts: reliability coef., separate or aggregated results
1000 professional actors
3 Studios, custom setup
200 TB of data shipped per week
20 behavioural actions per actor
Hollywood, California
Collected in: 4 months
900M annotated frames
1500 annotators
60k people recorded
–
25 countries
Collected in: 3 months
136k iPhone selfie-videos
5k people recorded
30 days recording
60GB per person
5 countries
Collected in: 4 months
Best suited for:
• Professional actors
• Multiple actions per actor
• Hollywood grade studios
Capabilities:
• Professionally directed Sessions
• ProRes, multi-angle, <1ms synchronicity
• Expert lighting, bespoke studio & hardware
• Sample up to 5K unique actors
• 50hrs p/w studio time
Best suited for:
• Smaller sample size
• Diverse or global sample
• Mobile recording locations
• Human guided participants
Capabilities:
• Directed recorded sessions
• Semi-pro equipment & set-up
• Sampling up to 100k participants
• 50 hrs / week per location
Best suited for:
• Large scale sample
• Diverse or global markets
• Long term engagement of participants
• Self-guided participants
Capabilities:
• Short and long-term engagements
• High quality recordings
• Automated QA validation
• Samples up to 100k respondents
Our network comprises of over 100,000 annotators drawn from a pool of more than 1 million addressable annotators worldwide to ensure a diverse level of expertise, cultural sensitivity and skill sets.
To uphold annotator quality, we employ a combination of proprietary Machine Learning Power and Human-In-The-Loop technology to ensure accuracy and reliability.
• 100,000+ qualified annotators
• Secure sensitive data protocols
• No software required
Our in-browser or app-based solutions enable data collection across 85 countries, accessing over 200 million people, with no or minimal software required.
• Access 85+ countries
• Up to 100k+ recordings per week
• Optimized for Android & iOS
• iPhone 3D depth data
• Video, audio and survey data
Vision AI is revolutionizing the gaming industry, elevating gaming immersion through lifelike non-player characters (NPCs), creating deeper emotional connections for richer player experiences.
Realeyes offers a range of solutions for collecting data, including partnering with professional directors, actors and producers for specific nuanced expression and gesture capture.
Get latest news and updates
