Visual Blocks: Visual prototyping of AI pipelines
summary
Visual Blocks is a browser visual programming platform that wires live webcam, video and audio through TensorFlow.js, Three.js and WebGL nodes with GPU shaders and supports Hugging Face transformer.js nodes both client-side and server task nodes. Fifteen ML practitioners cut prototyping from roughly one hour to minutes, case studies include portrait depth, scene fog and audio denoising, and InstructPipe auto-generates pipelines, reducing user interaction by 81%. Custom JavaScript nodes, interactive data augmentation, and shareable URLs are implemented and demonstrated working end to end.
AI-written summary · about
chapters (8)
- 0:00Intro
- 0:25Thank you
- 2:22Methodology
- 3:57Features
- 6:28New Nodes
- 8:47Custom Nodes
- 12:11Image prompt
- 13:07Conclusion
related talks
23mWebAI: Building the next user experience in the browser
24mWeb Apps of the Future With Web AI
24mThe next level apps for Web developers Machine learning on browsers 🤯
34mThe Browser, The Machine Learning and The Future we might not have imagined just less than a year ago
22mBoring but Effective: WebAI That Ships
25mJon Peck - SeattleJS Conference 2017
33mLocal Models: How to Run AI in the Browser
28mA visual programming language for animation, geospatial and the web