Example notebooks demonstrating how to use Pixeltable with Backblaze B2 Cloud Storage for multimodal data processing and AI workflows.
These examples show how to manage and process image and video data directly from B2, perform AI-driven transformations, and store the results back in cloud storage with automatic URL generation.
Learn how to extract video frames, transform them (e.g., grayscale conversion or AI-based editing), and store the results in Backblaze B2.
Pixeltable is an open-source AI data infrastructure that enables:
- Computed Columns: Automate data processing with AI models and transformations
- Multimodal Support: Work with images, video, audio, and documents in one framework
- Declarative Storage: Define where to store data and Pixeltable handles uploads and URL generation
- Persistent Storage: Maintain data and results across sessions
Learn more in the Pixeltable documentation.
Backblaze B2 Cloud Storage is cost-effective, S3-compatible cloud storage designed for simplicity and performance. It integrates seamlessly with Pixeltable using standard S3 endpoints (https://s3.{region}.backblazeb2.com/).
Key Benefits:
- S3-compatible API: Works seamlessly with Pixeltable’s storage system
- Cost-effective: Competitive pricing for scalable cloud storage
- Simple setup: Minimal configuration required
- Automatic URL generation: Pixeltable generates accessible URLs for stored outputs
Prerequisites: Backblaze B2 account (free tier available) and Python 3.9+