Machine Learning's Missed Opportunity in Visual Data Management
summary
ApertureData’s VDMS-based platform for visual data management: how it stores and indexes images and video along with evolving metadata, feature vectors and bounding boxes, supports similarity search and in-place preprocessing, and plugs into ML pipelines with multiple backends. The talk presents performance claims (up to 5x query improvement and large time savings for medical imaging ML deployments) and includes a demo showing how to use the platform.
AI-written summary · about
chapters (6)
- 0:00Start
- 1:05Current solutions are too complex and lack efficiency
- 2:01The intersection of ML + Big visual data is a huge unsolved problem
- 3:02ApertureData accelerates access and extraction operations across large visual data
- 11:24similarity searches
- 16:54Demo VIDEOS!
related talks
06mMachines and the magic of fast learning
33mArchitecting Scalable Multi-Modal Video Search
24mMachine Learning & Medical Imaging: The Future of Early Cancer Detection
28mVision, Big Data and AI
33mMachine Learning in Visual Search
42mApplications in ML Drug Discovery pipelines
33mTowards a deeper understanding of retrieval and vector databases
41mSearching for Products by Photo