The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence) Data-Centric Zero-Shot Learning for Precision Agriculture with Dimitris Zermas - #615
Feb 6, 2023
Dimitris Zermas, principal scientist at Sentera, shares insights on leveraging machine learning for precision agriculture. He discusses innovative tools like drones and cameras that enhance crop management. The conversation delves into challenges with plant counting and data imbalance, while also unveiling the power of zero-shot learning for efficient data use. Dimitris emphasizes a data-centric approach, detailing how strategic data selection can significantly reduce annotation time and costs, reshaping approaches in agricultural technology.
AI Snips
Chapters
Transcript
Episode notes
Early Plant Counting at Sentera
- Sentera initially used classical computer vision for plant counting.
- This proved challenging due to varying light, soil, and other environmental factors.
Deep Learning and Corner Cases
- Deep learning improved plant counting accuracy, but corner cases remained.
- These corner cases involved smaller plants, overlapping leaves, and distinguishing plant types.
Data Selection for Annotation
- Strategically select images for annotation to avoid unnecessary costs.
- Focus on images that add new information rather than overtraining on existing data.